Collaborative reasoning with multiple agents offers the potential for more robust and diverse problem-solving. However, existing approaches often suffer from homogeneous agent behaviors and lack of reflective and rethinking capabilities. We propose Multi-Agent Personality Shaping ((MAPS), a novel framework that enhances reasoning through agent diversity and internal critique. Inspired by the Big Five personality theory, MAPS assigns distinct personality traits to individual agents, shaping their reasoning styles and promoting heterogeneous collaboration. To enable deeper and more adaptive reasoning, MAPS introduces a Critic agent that reflects on intermediate outputs, revisits flawed steps, and guides iterative refinement. This integration of personality-driven agent design and structured collaboration improves both reasoning depth and flexibility. Empirical evaluations across three benchmarks demonstrate the strong performance of MAPS, with further analysis confirming its generalizability across different large language models and validating the benefits of multi-agent collaboration.

Large language models (LLMs) typically operate in a question-answering paradigm, where the quality of the input prompt critically affects the response. Automated Prompt Optimization (APO) aims to overcome the cognitive biases of manually crafted prompts and explore a broader prompt design space. However, existing APO methods often suffer from rigid template structures and inefficient exploration in the prompt space. To this end, we propose a Multi-Agent Adaptive Reasoning with Socratic guidance framework (MARS) for APO. MARS consists of five complementary agents and formulates the optimization process as a Partially Observable Markov Decision Process (POMDP), enabling adaptive prompt refinement through explicit state modeling and interactive feedback. Specifically, a Planner agent generates flexible optimization trajectories, a Teacher-Critic-Student triad engages in Socratic-style dialogue to iteratively optimize the prompt based on pseudo-gradient signals in the text space, and a Target agent executes the prompt in downstream tasks to provide performance feedback. MARS integrates reasoning, feedback, and state transition into a unified hidden-state evolution process, improving both the effectiveness and interpretability of optimization. Extensive experiments on multiple datasets demonstrate that MARS outperforms existing APO methods in terms of optimization performance, search efficiency, and interpretability.

Yumeng Fu, Jiayin Zhu, Lingling Zhang, Wenjun Wu, Bo Zhao, Shaoxuan Ma, Yushun Zhang, Jun Liu. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Yifei Li, Weidong Guo, Lingling Zhang, Rongman Xu, Muye Huang, Hui Liu, Lijiao Xu, Yu Xu, Jun Liu. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Hang Yan, Fangzhi Xu, Rongman Xu, Yifei Li, Jian Zhang, Haoran Luo, Xiaobao Wu, Anh Tuan Luu, Haiteng Zhao, Qika Lin, Jun Liu. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Xinyu Zhang, Yuchen Wan, Boxuan Zhang, Zesheng Yang, Lingling Zhang, Bifan Wei, Jun Liu. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Large language models (LLMs) have achieved remarkable success across a wide range of applications especially when augmented by external knowledge through retrieval-augmented generation (RAG). Despite their widespread adoption, recent studies have shown that LLMs often struggle to perform faithful reasoning when conflicting knowledge is retrieved. However, existing work primarily focuses on conflicts between external knowledge and the parametric knowledge of LLMs, leaving conflicts across external knowledge largely unexplored. Meanwhile, modern RAG systems increasingly emphasize the integration of unstructured text and (semi-)structured data like knowledge graphs (KGs) to improve knowledge completeness and reasoning faithfulness. To address this gap, we introduce ConflictQA, a novel benchmark that systematically instantiates conflicts between textual evidence and KG evidence. Extensive evaluations across representative LLMs reveal that, facing such cross-source conflicts, LLMs often fail to identify reliable evidence for correct reasoning. Instead, LLMs become more sensitive to prompting choices and tend to rely exclusively on either KG or textual evidence, resulting in incorrect responses. Based on these findings, we further propose XoT, a two-stage explanation-based thinking framework tailored for reasoning over heterogeneous conflicting evidence, and verify its effectiveness with extensive experiments.

Image-sentence matching that aims to understand the correspondence between vision and language, has achieved significant progress with various deep methods trained under large-scale supervision. Different from natural images taken by camera, diagrams in the textbooks contain more graphic objects, drawings, and natural objects, and the diagram-sentence matching plays an important role in textbook understanding and question answering. However, existing matching models are not suitable for the challenging task between diagrams and sentences, due to the more serious few-shot content and incomplete description problems. In this paper, we propose a novel local-feedback self-regulating memory framework (LFSRM) for diagram-sentence matching. On one hand, LFSRM includes an external memory to store the useful multi-modal information, especially uncommon ones, to overcome the few-shot content problem, where the memory is updated flexibly according to the local-feedback from visual-textual alignment scores. On the other hand, LFSRM designs an attention mechanism on local-level alignment scores and a strengthening factor impacted on sentence-to-diagram matching direction for alleviating the incomplete description problem. Extensive experiments on three datasets show that LFSRM achieves satisfactory results on conventional image-sentence matching, and outperforms SOTA methods on few-shot image/diagram-sentence matching by a large margin.

Geometry problem solving (GPS) requires high-level symbolic and logical reasoning based on geometry theorem knowledge to arrive at the answer. Despite the remarkable advances achieved by Large Language Models (LLMs) in various problem-solving tasks, they still struggle to perform rigorous multi-step geometry reasoning, which is essential for GPS. In this paper, we propose a dynamic tree-based geometry problem solver named GeoTree, which combines a knowledgeable LLM with a rigorous symbolic solver to perform geometry reasoning cooperatively. Specifically, an iterative multi-step geometry reasoning process is performed dynamically based on a tree-like structure, thereby emulating divergent and deliberate human problem-solving thinking. Each geometry reasoning step is completed collaboratively through four components, consisting ofTheorem Seeker,Symbolic Solver,Evaluator, andController. First,Theorem Seekerprompts LLMs to seek out candidate theorems with their inherent geometry theorem knowledge. Subsequently,Symbolic Solverapplies the theorems on the known conditions to obtain new additional conditions. Then,Evaluatorassesses the availability of the theorems and prompts LLMs to judge the usefulness of these new conditions for the problem target, which serves as the heuristic guidance for subsequent reasoning. Finally,Controllerdetermines the termination state, which decides whether to continue invoking the other three components for further attempts. Extensive experiments on Geometry3K demonstrate the superiority of GeoTree in accuracy, efficiency, and explainability.

Chart understanding enables automated data analysis for humans, which requires models to achieve highly accurate visual comprehension. While existing Visual Language Models (VLMs) have shown progress in chart understanding, the lack of high-quality training data and comprehensive evaluation benchmarks hinders VLM chart comprehension. In this paper, we introduce EvoChart, a novel self-training method for generating synthetic chart data to enhance VLMs' capabilities in real-world chart comprehension. We also propose EvoChart-QA, a noval benchmark for measuring models' chart comprehension abilities in real-world scenarios. Specifically, EvoChart is a unique self-training data synthesis approach that simultaneously produces high-quality training corpus and a high-performance chart understanding model. EvoChart-QA consists of 650 distinct real-world charts collected from 140 different websites and 1,250 expert-curated questions that focus on chart understanding. Experimental results on various open-source and proprietary VLMs tested on EvoChart-QA demonstrate that even the best proprietary model, GPT-4o, achieves only 49.8% accuracy. Moreover, the EvoChart method significantly boosts the performance of open-source VLMs on real-world chart understanding tasks, achieving 54.2% accuracy on EvoChart-QA.

Charts are widely used for data visualization across various fields, including education, research, and business. Chart Question Answering (CQA) is an emerging task focused on the automatic interpretation and reasoning of data presented in charts. However, chart images are inherently difficult to interpret, and chart-related questions often involve complex logical and numerical reasoning, which hinders the performance of existing models. This paper introduces VProChart, a novel framework designed to address these challenges in CQA by integrating a lightweight Visual Perception Alignment Agent (VPAgent) and a Programmatic Solution Reasoning approach. VPAgent aligns and models chart elements based on principles of human visual perception, enhancing the understanding of chart context. The Programmatic Solution Reasoning approach leverages large language models (LLMs) to transform natural language reasoning questions into structured solution programs, facilitating precise numerical and logical reasoning. Extensive experiments on benchmark datasets such as ChartQA and PlotQA demonstrate that VProChart significantly outperforms existing methods, highlighting its capability in understanding and reasoning with charts.

Large Language Models (LLMs) may suffer from hallucinations in real-world applications due to the lack of relevant knowledge. In contrast, knowledge graphs encompass extensive, multi-relational structures that store a vast array of symbolic facts. Consequently, integrating LLMs with knowledge graphs has been extensively explored, with Knowledge Graph Question Answering (KGQA) serving as a critical touchstone for the integration. This task requires LLMs to answer natural language questions by retrieving relevant triples from knowledge graphs. However, existing methods face two significant challenges: *excessively long reasoning paths distracting from the answer generation*, and *false-positive relations hindering the path refinement*. In this paper, we propose an iterative interactive KGQA framework that leverages the interactive learning capabilities of LLMs to perform reasoning and Debating over Graphs (DoG). Specifically, DoG employs a subgraph-focusing mechanism, allowing LLMs to perform answer trying after each reasoning step, thereby mitigating the impact of lengthy reasoning paths. On the other hand, DoG utilizes a multi-role debate team to gradually simplify complex questions, reducing the influence of false-positive relations. This debate mechanism ensures the reliability of the reasoning process. Experimental results on five public datasets demonstrate the effectiveness and superiority of our architecture. Notably, DoG outperforms the state-of-the-art method ToG by 23.7% and 9.1% in accuracy on WebQuestions and GrailQA, respectively. Furthermore, the integration experiments with various LLMs on the mentioned datasets highlight the flexibility of DoG.

In this work, we address the challenging task of Generalized Referring Expression Comprehension (GREC). Compared to the classic Referring Expression Comprehension (REC) that focuses on single-target expressions, GREC extends the scope to a more practical setting by further encompassing no-target and multi-target expressions. Existing REC methods face challenges in handling the complex cases encountered in GREC, primarily due to their fixed output and limitations in multi-modal representations. To address these issues, we propose a Hierarchical Alignment-enhanced Adaptive Grounding Network (HieA2G) for GREC, which can flexibly deal with various types of referring expressions. First, a Hierarchical Multi-modal Semantic Alignment (HMSA) module is proposed to incorporate three levels of alignments, including word-object, phrase-object, and text-image alignment. It enables hierarchical cross-modal interactions across multiple levels to achieve comprehensive and robust multi-modal understanding, greatly enhancing grounding ability for complex cases. Then, to address the varying number of target objects in GREC, we introduce an Adaptive Grounding Counter (AGC) to dynamically determine the number of output targets. Additionally, an auxiliary contrastive loss is employed in AGC to enhance object-counting ability by pulling in multi-modal features with the same counting and pushing away those with different counting. Extensive experimental results show that HieA2G achieves new state-of-the-art performance on the challenging GREC task and also the other 4 tasks, including REC, Phrase Grounding, Referring Expression Segmentation (RES), and Generalized Referring Expression Segmentation (GRES), demonstrating the remarkable superiority and generalizability of the proposed HieA2G.

Fangzhi Xu, Hang Yan, Chang Ma, Haiteng Zhao, Qiushi Sun, Kanzhi Cheng, Junxian He, Jun Liu, Zhiyong Wu. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

One of the primary driving forces contributing to the superior performance of Large Language Models (LLMs) is the extensive availability of human-annotated natural language data, which is used for alignment fine-tuning. This inspired researchers to investigate self-training methods to mitigate the extensive reliance on human annotations. However, the current success of self-training has been primarily observed in natural language scenarios, rather than in the increasingly important neural-symbolic scenarios. To this end, we propose an environment-guided neural-symbolic self-training framework named ENVISIONS. It aims to overcome two main challenges: (1) the scarcity of symbolic data, and (2) the limited proficiency of LLMs in processing symbolic language. Extensive evaluations conducted on three distinct domains demonstrate the effectiveness of our approach. Additionally, we have conducted a comprehensive analysis to uncover the factors contributing to ENVISIONS’s success, thereby offering valuable insights for future research in this area.

Xinyu Zhang, Lingling Zhang, Yanrui Wu, Muye Huang, Wenjun Wu, Bo Li, Shaowei Wang, Basura Fernando, Jun Liu. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

Considerable attention has been paid to predicting student performance on exercises. The performance of prior studies is determined by the quality of the trait features of students and exercises. Nevertheless, most of the prior study primarily examines simple pairwise interactions in learning trait features, like those between students and exercises or exercises and concepts, while disregarding the complex higher-order interactions that typically exist among these components, which in turn hinders the prediction results. In this paper, we using an innovative Multi-Channel Graph Contrastive Learning (MCGCL) framework that integrates various high-order interactions for predicting student performance. MCGCL characterizes graph structures reflecting various high-order relationships among students, exercises, and concepts through multiple channels, thereby enhancing the trait features of both students and exercises. Moreover, graph contrastive learning is employed to enhance the representation of trait features acquired from high-order graph structures in diverse views. Extensive experiments on real-world datasets show that MCGCL achieves state-of-the-art results on the task of predicting student performance. The code is available at https://github.com/sunlitsong/MCGCL.

Traditional knowledge graph (KG) embedding methods aim to represent entities and relations in a low-dimensional space, primarily focusing on static graphs. However, real-world KGs are dynamically evolving with the constant addition of entities, relations and facts. To address such dynamic nature of KGs, several continual knowledge graph embedding (CKGE) methods have been developed to efficiently update KG embeddings to accommodate new facts while maintaining learned knowledge. As KGs grow at different rates and scales in real-world scenarios, existing CKGE methods often fail to consider the varying scales of updates and lack systematic evaluation throughout the entire update process. In this paper, we propose SAGE, a scale-aware gradual evolution framework for CKGE. Specifically, SAGE firstly determine the embedding dimensions based on the update scales and expand the embedding space accordingly. The Dynamic Distillation mechanism is further employed to balance the preservation of learned knowledge and the incorporation of new facts. We conduct extensive experiments on seven benchmarks, and the results show that SAGE consistently outperforms existing baselines, with a notable improvement of 1.38% in MRR, 1.25% in H@1 and 1.6% in H@10. Furthermore, experiments comparing with fixed dimensions methods show that SAGE achieves optimal performance on every snapshot, demonstrating the importance of adaptive embedding dimensions in CKGE.

Abstract visual reasoning, exemplified by Raven's Progressive Matrices (RPM), remains a significant challenge in artificial intelligence. A critical difficulty lies in disentangling abstract relational rules from image-specific features, as these rules operate independently of visual appearances. To address this challenge, we propose the Cognitive Predictive Coding Network (CPCN), inspired by predictive coding theory from cognitive science. CPCN features a three-component architecture: a Relation Disentangler that separates abstract rules from image-specific features through prediction error minimization; Stacked Free Energy Minimizers (FEMs) that leverage energy minimization principles to progressively reduce uncertainty during hierarchical abstraction; and a classifier for solution identification. Unlike previous approaches, our model employs mutual information constraints to explicitly separate relation-relevant and relation-irrelevant features, enabling more robust pattern recognition. Our novel FEMs provide a principled approach to uncertainty reduction through iterative refinement of pattern understanding. Experiments demonstrate PAH's superior performance across multiple benchmarks (such as 98.9% on RAVEN-Fair) with state-of-the-art 59.7% average accuracy across all PGM subtasks.

Charts are high-density visualization carriers for complex data, serving as a crucial medium for information extraction and analysis. Automated chart understanding poses significant challenges to existing multimodal large language models (MLLMs) due to the need for precise and complex visual reasoning. Current step-by-step reasoning models primarily focus on text-based logical reasoning for chart understanding. However, they struggle to refine or correct their reasoning when errors stem from flawed visual understanding, as they lack the ability to leverage multimodal interaction for deeper comprehension. Inspired by human cognitive behavior, we propose ChartSketcher, a multimodal feedback-driven step-by-step reasoning method designed to address these limitations. ChartSketcher is a chart understanding model that employs Sketch-CoT, enabling MLLMs to annotate intermediate reasoning steps directly onto charts using a programmatic sketching library, iteratively feeding these visual annotations back into the reasoning process. This mechanism enables the model to visually ground its reasoning and refine its understanding over multiple steps. We employ a two-stage training strategy: a cold start phase to learn sketch-based reasoning patterns, followed by off-policy reinforcement learning to enhance reflection and generalization. Experiments demonstrate that ChartSketcher achieves promising performance on chart understanding benchmarks and general vision tasks, providing an interactive and interpretable approach to chart comprehension.

Knowledge graph-based retrieval-augmented generation seeks to mitigate hallucinations in Large Language Models (LLMs) caused by insufficient or outdated knowledge. However, existing methods often fail to fully exploit the prior knowledge embedded in knowledge graphs (KGs), particularly their structural information and explicit or implicit constraints. The former can enhance the faithfulness of LLMs' reasoning, while the latter can improve the reliability of response generations. Motivated by these, we propose a trustworthy reasoning framework, termed Deliberation over Priors (\texttt{DP}), which sufficiently utilizes the priors contained in KGs. Specifically, \texttt{DP} adopts a progressive knowledge distillation strategy that integrates structural priors into LLMs through a combination of supervised fine-tuning and Kahneman-Tversky Optimization, thereby improving the faithfulness of relation path generation. Furthermore, our framework employs a reasoning-introspection strategy, which guides LLMs to perform refined reasoning verification based on extracted constraint priors, ensuring the reliability of response generation. Extensive experiments on three benchmark datasets demonstrate that \texttt{DP} achieves new state-of-the-art performance, especially a H@1 improvement of 13% on the ComplexWebQuestions dataset, and generates highly trustworthy responses. We also conduct various analyses to verify its flexibility and practicality. Code is available at [https://github.com/mira-ai-lab/Deliberation-on-Priors](https://github.com/mira-ai-lab/Deliberation-on-Priors).

The task of geometry problem solving has been a long-standing focus in the automated mathematics community and draws growing attention due to its complexity for both symbolic and neural models. Although prior studies have explored various effective approaches for enhancing problem solving performances, two fundamental challenges remain unaddressed, which are essential to the application in practical scenarios. First, the multi-step reasoning gap between the initial geometric conditions and ultimate problem goal leads to a great search space for solution exploration. Second, obtaining multiple interpretable and shorter solutions remains an open problem. In this work, we introduce the Causal-Reasoning Geometry Problem Solver to overcome these challenges. Specifically, the Causal Graph Reasoning theory is proposed to perform symbolic reasoning before problem solving. Several causal graphs are constructed according to predefined rule base, where each graph is composed of primitive nodes, causal edges and prerequisite edges. By applying causal graph deduction from initial conditions, the reachability status of nodes are iteratively conveyed by causal edges until reaching the target nodes, representing feasible causal deduction paths. In this way, the search space of solutions is compressed from the beginning, the end and intermediate reasoning paths, while ensuring the interpretability and variety of solutions. To achieve this, we further propose Forward Matrix Deduction which transforms the causal graphs into matrices and vectors, and applies matrix operations to update the status value of reachable nodes in iterations. Finally, multiple solutions can be generated by tracing back from the target nodes after validation. Experiments demonstrate the effectiveness of our method to obtain multiple shorter and interpretable solutions. Code is available after acceptance.

Diagram question answering (DQA) is a challenging task that requires models to combine with domain-specific knowledge and reason over the diagrams to answer questions. Multimodal Large Language Models (MLLMs) have recently made notable strides in combining textual and visual information, emerging as a promising solution for addressing the DQA task. However, they still encounter challenges in deliberate multimodal reasoning over the fine-grained visual details of content-rich and knowledge-grounded diagrams. The tight interweaving of visual and textual reasoning for MLLMs is also susceptible to hallucinations. To overcome these limitations, we propose a Glance-then-Focus Multimodal Reasoning framework named GlFoMR for DQA, which features a flexible architecture for comprehensive visual and text interaction. Firstly, the diagram is parsed into a hierarchical structure spanning different granularities including isolated single-object, object-group, and whole-diagram. Subsequently, the Glance-Plan and Focus-Reason stages collaborate to decouple the complex reasoning process. Glance-Plan first generates a preliminary plan by glancing at the multimodal context, specifying sub-goals related to knowledge extraction, visual perception, and visual reasoning. Based on these sub-goals, Focus-Reason further integrates domain-specific knowledge and visual details to enable more deliberate reasoning. The parsed multi-granularity diagram information is seamlessly incorporated into the corresponding sub-goal achievement process, enhancing the perception and reasoning capabilities of MLLMs for better DQA performance. Extensive experimental results on four DQA datasets demonstrate that GlFoMR achieves substantial improvements, showcasing its potential to advance the development of multimodal reasoning.

Continual knowledge graph embedding (CKGE) has gained wide attention for managing dynamic knowledge graphs (KGs), which are continuously updated with new facts. Unlike traditional methods designed for static KGs, CKGE enables incremental updates to KG embeddings to accommodate new facts while retaining previously learned knowledge. Despite these advancements, current CKGE studies and benchmarks primarily focus on handling the increasing scale of data while overlooking changes in graph patterns. These changes, altering the graph structure of KGs, are referred to as pattern shifts in this paper. Pattern shifts frequently arise as new facts are added, introducing significant challenges to the stability and adaptability of CKGE methods. To address this gap, we introduce a suite of novel and challenging benchmarks, called PS-CKGE, specifically designed to evaluate CKGE methods under pattern shifts, where logic rules are utilized to capture and manage structural changes in dynamic KGs. Through these benchmarks, we comprehensively evaluate current CKGE methods in terms of their overall performance, resistance to catastrophic forgetting, and adaptability to new knowledge. The results show that pattern shifts not only exacerbate their risk of catastrophic forgetting but also impair their adaptability, usually with greater performance degradation over triples associated with more significant changes.

Knowledge Graph (KG) reasoning aims to predict missing entities in incomplete triples, which requires adequate structural information to derive accurate embeddings. However, KGs in the real world are not as dense as the idealized benchmarks, where sparse graph structures restrict the comprehensive structural information for superior performance. Although the logical semantics in KGs shows its potential in alleviating the impact of structural sparsity, there still exist some challenges. The deficient supervision and the semantic gap of logic make it difficult to introduce logical semantics in sparse KG reasoning. To this end, we propose a novel KG reasoning approach LoLLM injecting logic with the supervised information supplied by the Large Language Model (LLM), which is proved to be effective in evaluating and scoring. Firstly, LoLLM derives structural embeddings employing a graph convolutional network (GCN) with relation-aware and triple-aware attention. LoLLM secondly constructs reasoning paths instantiated from the first-order logic rules extracted from sparse KGs, and injects the logical semantics by a designed LLM-enhanced tuning strategy. We propose a textual loss (TL) and a logical loss (LL) in the optimization and obtain logical tuning embeddings of KG in this process. Finally, LoLLM fuses structural embeddings from the GCN and logical tuning embeddings from the LLM-enhanced tuning for scoring and incomplete triple prediction. Extensive experiments on two sparse KGs and a benchmark show that LoLLM outperforms state-of-the-art structure-based and Language Model (LM)-augmented baselines. Moreover, the logic rules with corresponding confidences provide explicit explanations as an interpretable paradigm.

Document-level relation extraction (DocRE) aims at predicting relations of all entity pairs in one document, which plays an important role in information extraction. DocRE is more challenging than previous sentence-level relation extraction, as it often requires coreference and logical reasoning across multiple sentences. Graph-based methods are the mainstream solution to this complex reasoning in DocRE. They generally construct the heterogeneous graphs with entities, mentions, and sentences as nodes, co-occurrence and co-reference relations as edges. Their performance is difficult to further break through because the semantics and direction of the relation are not jointly considered in graph inference process. To this end, we propose a novel translation-guided double-graph inference network named TDGI for DocRE. On one hand, TDGI includes two relation semantics-aware and direction-aware reasoning graphs, i.e., mention graph and entity graph, to mine relations among long-distance entities more explicitly. Each graph consists of three elements: vectorized nodes, edges, and direction weights. On the other hand, we devise an interesting translation-based graph updating strategy that guides the embeddings of mention/entity nodes, relation edges, and direction weights following the specific translation algebraic structure, thereby to enhance the reasoning skills of TDGI. In the training procedure of TDGI, we minimize the relation multi-classification loss and triple contrastive loss together to guarantee the model's stability and robustness. Comprehensive experiments on three widely-used datasets show that TDGI achieves outstanding performance comparing with state-of-the-art baselines.

Memory-based Temporal Graph Neural Networks (M-TGNNs) demonstrate superior performance in dynamic graph learning tasks. Their success attributes to a memory module, which captures historical information for each node and implicitly creates a memory dependency constraint among chronologically ordered minibatches. This unique characteristic of M-TGNN introduces new challenges for parallel training that have not been encountered before. Existing parallelism strategies for M-TGNN either sacrifice memory accuracy (minibatch parallelism and epoch parallelism) or compromise space efficiency (memory parallelism) to optimize runtime. This paper proposes a pipeline parallel approach for multi-GPU M-TGNN training that effectively addresses both inter-minibatch memory dependencies and intra-minibatch task dependencies, based on a runtime analysis DAG for M-TGNNs. We further optimize pipeline efficiency by incorporating improved scheduling, finer-grained operation reorganization, and targeted communication optimizations tailored to the specific training properties of M-TGNN. These enhancements significantly reduce GPU waiting and idle time caused by memory dependencies and frequent communication and result in zero pipeline bubbles for common training configurations. Extensive evaluations demonstrate that PipeTGL achieves a speedup of 1.27x to 4.74x over other baselines while also improving the accuracy of M-TGNN training across multiple GPUs.

Tabular data is prevalent in many fields. In practice, tabular data classification may encounter severe challenges due to class imbalance, i.e., some majority classes overwhelm minority ones. Such imbalance could lead to biased prediction tendency of trained classifiers towards majority classes. Oversampling minority classes is an essential solution due to its generality and independence of downstream tasks. Recent years have witnessed the advantages of generative adversarial networks (GANs) in synthetic data generation, favored for their ability to generate quasi-realistic samples. However, challenges arise when the size of minority classes is too small to provide sufficient information for learning real data distributions. Furthermore, the generated minority-class samples could exacerbate the class overlap problem, i.e., some generated samples unexpectedly overlap with partial majority-class samples. To address these challenges, this paper presents B2BGAN, a novel GAN-based approach for oversampling imbalanced tabular data. To capture the real data distribution in a fine-grained manner, we propose a novel backbone-to-branches neural network for the generator to fit the majority and minority classes simultaneously. The backbone network fits the whole distribution of the entire data, while each branch network grasps the distinctive characteristics of individual classes. To alleviate the class overlap problem of generated samples, we develop a prototype-guided loss function to ensure that generated samples are closer to the corresponding class prototypes. We evaluate the effectiveness of B2BGAN on six real-world datasets using six metrics. Experimental results demonstrate that our method outperforms state-of-the-art models by 5.38% in AUC and 10.19% in AP.

Logical reasoning consistently plays a fundamental and significant role in the domains of knowledge engineering and artificial intelligence. Recently, Large Language Models (LLMs) have emerged as a noteworthy innovation in natural language processing (NLP). However, the question of whether LLMs can effectively address the task of logical reasoning, which requires gradual cognitive inference similar to human intelligence, remains unanswered. To this end, we aim to bridge this gap and provide comprehensive evaluations in this paper. First, to offer systematic evaluations, we select fifteen typical logical reasoning datasets and organize them into deductive, inductive, abductive and mixed-form reasoning settings. Considering the comprehensiveness of evaluations, we include 3 early-era representative LLMs and 4 trending LLMs. Second, different from previous evaluations relying only on simple metrics (e.g.,accuracy), we propose fine-level evaluations in objective and subjective manners, covering both answers and explanations, includinganswer correctness,explain correctness,explain completenessandexplain redundancy. Additionally, to uncover the logical flaws of LLMs, problematic cases will be attributed to five error types from two dimensions, i.e.,evidence selection processandreasoning process. Third, to avoid the influences of knowledge bias and concentrate purely on benchmarking the logical reasoning capability of LLMs, we propose a new dataset with neutral content. Based on the in-depth evaluations, this paper finally forms a general evaluation scheme of logical reasoning capability from six dimensions (i.e.,Correct,Rigorous,Self-aware,Active,OrientedandNo hallucination). It reflects the pros and cons of LLMs and gives guiding directions for future works.

Diagram question answering (DQA), which is defined as answering natural language questions according to the visual diagram context, has attracted attention and has recently become a new benchmark for evaluating the complex reasoning ability of models. However, this reasoning task is extremely challenging because of the inclusion of abstract visual objects and specialized textual terms, as well as the complex relationships between them. The rarity of data caused by the high cost of annotation also makes large-scale deep models invalid for the DQA task. To address the above challenges, this paper proposes the cross-modal alignment-guided self-supervised learning model for DQA (CAS-DQA). Unlike previous works, the CAS-DQA model focuses on learning internal visual-textual object relationships, innovatively proposes an attention mechanism module based on object alignment, and effectively integrates cross-modal knowledge units for diagram understanding. In addition, the CAS-DQA model constructs two self-supervised learning (SSL) tasks via intermediate results of visual-textual object alignment. These two tasks exploit the unnoticed objects inside the diagram to fully and completely understand the diagram. They also effectively increase the amount of diagram question-answering data to address the challenge of data scarcity. To the best of our knowledge, the CAS-DQA model is the first to extend SSL strategies to the diagram question-answering task. We evaluate the CAS-DQA model on three different datasets. The results of extensive experiments show that our model significantly outperforms baselines on different scenarios and that the internal object alignment module and self-supervised tasks produce excellent results.

Cross-modal retrieval (CMR) aims to establish interaction between different modalities, among which supervised CMR is emerging due to its flexibility in learning semantic category discrimination. Despite the remarkable performance of previous supervised CMR methods, much of their success can be attributed to the well-annotated data. However, even for unimodal data, precise annotation is expensive and time-consuming, and it becomes more challenging with the multimodal scenario. In practice, massive multimodal data are collected from the Internet with coarse annotation, which inevitably introduces noisy labels. Training with such misleading labels would bring two key challenges-enforcing the multimodal samples to align incorrect semantics and widen the heterogeneous gap, resulting in poor retrieval performance. To tackle these challenges, this work proposes UOT-RCL, a unified framework based on optimal transport (OT) for robust CMR. First, we propose a semantic alignment based on partial OT to progressively correct the noisy labels, where a novel cross-modal consistent cost function is designed to blend different modalities and provide precise transport cost. Second, to narrow the discrepancy in multimodal data, an OT-based relation alignment is proposed to infer the semantic-level cross-modal matching. Both of these components leverage the inherent correlation among multimodal data to facilitate effective cost function. The experiments on three widely used CMR datasets demonstrate that our UOT-RCL surpasses the state-of-the-art approaches and significantly improves the robustness against noisy labels.

Current clustering-based open relation extraction (OpenRE) methods usually apply clustering algorithms on top of pre-trained language models.However, this practice has three drawbacks.First, embeddings from language models are high-dimensional and anisotropic, so using simple metrics to calculate distances between these embeddings may not accurately reflect the relational similarity.Second, there exists a gap between the pre-trained language models and downstream clustering for their different objective forms.Third, clustering with embeddings deviates from the primary aim of relation extraction, as it does not directly obtain relations.In this work, we propose a new idea for OpenRE in the era of LLMs, that is, extracting relational phrases and directly exploiting the knowledge in LLMs to assess the semantic similarity between phrases without relying on any additional metrics.Based on this idea, we developed a framework, ORELLM, that makes two LLMs work collaboratively to achieve clustering and address the above issues.Experimental results on different datasets show that ORELLM outperforms current baselines by 1.4% ∼ 3.13% in terms of clustering accuracy.

Logical reasoning task has attracted great interest since it was proposed.Faced with such a task, current competitive models, even large language models (e.g., ChatGPT and PaLM 2), still perform badly.Previous promising LMs struggle in logical consistency modeling and logical structure perception.To this end, we model the logical reasoning task by transforming each logical sample into reasoning paths and propose an architecture PathReasoner.It addresses the task from the views of both data and model.To expand the diversity of the logical samples, we propose an atom extension strategy supported by equivalent logical formulas, to form new reasoning paths.From the model perspective, we design a stack of transformer-style blocks.In particular, we propose a path-attention module to joint model in-atom and cross-atom relations with the highorder diffusion strategy.Experiments show that PathReasoner achieves competitive performances on two logical reasoning benchmarks and great generalization abilities.

Fangzhi Xu, Zhiyong Wu, Qiushi Sun, Siyu Ren, Fei Yuan, Shuai Yuan, Qika Lin, Yu Qiao, Jun Liu. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

Document-level Event Argument Extraction (DEAE) aims to identify arguments and their specific roles from an unstructured document. The advanced approaches on DEAE utilize prompt-based methods to guide pre-trained language models (PLMs) in extracting arguments from input documents. They mainly concentrate on establishing relations between triggers and entity mentions within documents, leaving two unresolved problems: a) independent modeling of entity mentions; b) document-prompt isolation. To this end, we propose a semantic mention Graph Augmented Model (GAM) to address these two problems in this paper. Firstly, GAM constructs a semantic mention graph that captures relations within and between documents and prompts, encompassing co-existence, co-reference and co-type relations. Furthermore, we introduce an ensemble graph transformer module to address mentions and their three semantic relations effectively. Later, the graph-augmented encoder-decoder module incorporates the relation-specific graph into the input embedding of PLMs and optimizes the encoder section with topology information, enhancing the relations comprehensively. Extensive experiments on the RAMS and WikiEvents datasets demonstrate the effectiveness of our approach, surpassing baseline methods and achieving a new state-of-the-art performance.

Diagram Question Answering (DQA) is a challenging task, requiring models to answer natural language questions based on visual diagram contexts. It serves as a crucial basis for academic tutoring, technical support, and more practical applications. DQA poses significant challenges, such as the demand for domain-specific knowledge and the scarcity of annotated data, which restrict the applicability of large-scale deep models. Previous approaches have explored external knowledge integration through pretraining, but these methods are costly and can be limited by domain disparities. While Large Language Models (LLMs) show promise in question-answering, there is still a gap in how to cooperate and interact with the diagram parsing process. In this paper, we introduce the Chain-of-Guiding Learning Model for Diagram Question Answering (CoG-DQA), a novel framework that effectively addresses DQA challenges. CoG-DQA leverages LLMs to guide diagram parsing tools (DPTs) through the guiding chains, enhancing the precision of diagram parsing while introducing rich background knowledge. Our experimental findings reveal that CoG-DQA surpasses all comparison models in various DQA scenarios, achieving an average accuracy enhancement exceeding 5% and peaking at 11% across four datasets. These results underscore CoG-DQA's capacity to advance the field of visual question answering and promote the integration of LLMs into specialized domains.

Geometry Problem Solving has drawn growing attention recently due to its application prospects in intelligent ed-ucation field. However, existing methods are still inade-quate to meet the needs of practical application, suffering from the following limitations: 1) explainability is not en-sured which is essential in real teaching scenarios; 2) the small scale and incomplete annotation of existing datasets make it hard for model to comprehend geometric knowl-edge. To tackle the above problems, we propose a novel method called Explainable Geometry Problem Solving (E-GPS). E-GPS first parses the geometric diagram and prob-lem text into unified formal language representations. Then, the answer and explainable reasoning and solving steps are obtained by a Top-Down Problem Solver (TD-PS), which innovatively solves the problem from the target and focuses on what is needed. To alleviate the data issues, a Bottom-Up Problem Generator (BU-PG) is devised to augment the data set with various well-annotated constructed geome-try problems. It enables us to train an enhanced theorem predictor with a better grasp of theorem knowledge, which further improves the efficiency ofTD-PS. Extensive experi-ments demonstrate that E-GPS maintains comparable solving performances with fewer steps and provides outstanding explainability.

Automatically generated questions often suffer from problems such as unclear expression or factual inaccuracies, requiring a reliable and comprehensive evaluation of their quality.Human evaluation is widely used in the field of question generation (QG) and serves as the gold standard for automatic metrics.However, there is a lack of unified human evaluation criteria, which hampers consistent and reliable evaluations of both QG models and automatic metrics.To address this, we propose QGEval, a multi-dimensional Evaluation benchmark for Question Generation, which evaluates both generated questions and existing automatic metrics across 7 dimensions: fluency, clarity, conciseness, relevance, consistency, answerability, and answer consistency.We demonstrate the appropriateness of these dimensions by examining their correlations and distinctions.Through consistent evaluations of QG models and automatic metrics with QGEval, we find that 1) most QG models perform unsatisfactorily in terms of answerability and answer consistency, and 2) existing metrics fail to align well with human judgments when evaluating generated questions across the 7 dimensions.We expect this work to foster the development of both QG technologies and their evaluation.

Embedding symbolic logical formulas into a low-dimensional continuous space provides an effective way for the Neural-Symbolic system. However, current studies are all constrained by the syntactic structure modeling and fail to preserve intrinsic semantics. To this end, we propose a novel model of Contrastive Graph Representations (ConGR) for logical formulas embedding. Firstly, it introduces a densely connected graph convolutional network (GCN) with an attention mechanism to process syntax parsing graphs of formulas. Secondly, the contrastive instances for each anchor formula are generated by the transformation under the guidance of logical properties. Two types of contrast, global-local and global-global, are carried out to refine formula embeddings with semantic information. Extensive experiments demonstrate that ConGR obtains superior performance against state-of-the-art baselines.

Feature-only partition of large graph data in distributed Graph Neural Network (GNN) training offers advantages over commonly adopted graph structure partition, such as minimal graph preprocessing cost and elimination of cross-worker subgraph sampling burdens. Nonetheless, performance bottleneck of GNN training with feature-only partitions still largely lies in the substantial communication overhead due to cross-worker feature fetching. To reduce the communication overhead and expedite distributed training, we first investigate and answer two key questions on convergence behaviors of GNN model in feature-partition based distribute GNN training: 1) As no worker holds a complete copy of each feature, can gradient exchange among workers compensate for the information loss due to incomplete local features? 2) If the answer to the first question is negative, is feature fetching in every training iteration of the GNN model necessary to ensure model convergence? Based on our theoretical findings on these questions, we derive an optimal communication plan that decides the frequency for feature fetching during the training process, taking into account bandwidth levels among workers and striking a balance between model loss and training time. Extensive evaluation demonstrates consistent results with our theoretical analysis, and the effectiveness of our proposed design.

Audio-Visual Question Answering (AVQA) is a complex multi-modal reasoning task, demanding intelligent systems to accurately respond to natural language queries based on audio-video input pairs. Nevertheless, prevalent AVQA approaches are prone to overlearning dataset biases, resulting in poor robustness. Furthermore, current datasets may not provide a precise diagnostic for these methods. To tackle these challenges, firstly, we propose a novel dataset, *MUSIC-AVQA-R*, crafted in two steps: rephrasing questions within the test split of a public dataset (*MUSIC-AVQA*) and subsequently introducing distribution shifts to split questions. The former leads to a large, diverse test space, while the latter results in a comprehensive robustness evaluation on rare, frequent, and overall questions. Secondly, we propose a robust architecture that utilizes a multifaceted cycle collaborative debiasing strategy to overcome bias learning. Experimental results show that this architecture achieves state-of-the-art performance on MUSIC-AVQA-R, notably obtaining a significant improvement of 9.32\%. Extensive ablation experiments are conducted on the two datasets mentioned to analyze the component effectiveness within the debiasing strategy. Additionally, we highlight the limited robustness of existing multi-modal QA methods through the evaluation on our dataset. We also conduct experiments combining various baselines with our proposed strategy on two datasets to verify its plug-and-play capability. Our dataset and code are available at <https://github.com/reml-group/MUSIC-AVQA-R>.

Knowledge graph embedding (KGE) methods have achieved great success in handling various knowledge graph (KG) downstream tasks. However, KGE methods may learn biased representations on low-quality KGs that are prevalent in the real world. Some recent studies propose adversarial attacks to investigate the vulnerabilities of KGE methods, but their attackers are target-oriented with the KGE method and the target triples to predict are given in advance, which lacks practicability. In this work, we explore untargeted attacks with the aim of reducing the global performances of KGE methods over a set of unknown test triples and conducting systematic analyses on KGE robustness. Considering logic rules can effectively summarize the global structure of a KG, we develop rule-based attack strategies to enhance the attack efficiency. In particular, we consider adversarial deletion which learns rules, applying the rules to score triple importance and delete important triples, and adversarial addition which corrupts the learned rules and applies them for negative triples as perturbations. Extensive experiments on two datasets over three representative classes of KGE methods demonstrate the effectiveness of our proposed untargeted attacks in diminishing the link prediction results. And we also find that different KGE methods exhibit different robustness to untargeted attacks. For example, the robustness of methods engaged with graph neural networks and logic rules depends on the density of the graph. But rule-based methods like NCRL are easily affected by adversarial addition attacks to capture negative rules.

Relation prediction in knowledge graphs (KGs) aims at predicting missing relations in incomplete triples, whereas the dominant paradigm by KG embeddings has a limitation to predict the relation between unseen entities. This situation is called an inductive setting, which is more common in the real-world scenario. To handle this issue, implicit symbolic rules have shown great potential in capturing the inductive capability. However, it is still challenging to obtain precise representations of logic rules from KGs. The argument variability and predicate non-commutativity in symbolic rule integration make the modeling of component symbols difficult. To this end, we propose a novel inductive relation prediction model named SymRITa with a logic transformer integrating rules. SymRITa firstly extracts the subgraph, whose embeddings are captured by a graph network. Meanwhile, symbolic rule graphs in the subgraph can be generated. Then, the symbolic rules are modeled by a proposed logic transformer. Specifically, the input format based on the subgraph-based embeddings is to focus on the argument variability in symbolic rules. In addition, a conjunction attention mechanism in the logic transformer can resolve predicate non-commutativity in the symbolic rule integration process. Finally, the subgraph-based and symbol-based embeddings obtained from the previous steps are combined for the training regime, and prediction results as well as rules explaining the reasoning process are explicitly output. Extensive experiments on twelve inductive datasets show that SymRITa achieves outstanding effectiveness compared to state-of-the-art inductive baselines. Moreover, the logic rules with corresponding confidences provide an interpretable paradigm.

Visual question answering requires a system to provide an accurate natural language answer given an image and a natural language question. However, it is widely recognized that previous generic VQA methods often tend to memorize biases present in the training data rather than learning proper behaviors, such as grounding images before predicting answers. Therefore, these methods usually achieve high in-distribution but poor out-of-distribution performance. In recent years, various datasets and debiasing methods have been proposed to evaluate and enhance the VQA robustness, respectively. This paper provides the first comprehensive survey focused on this emerging fashion. Specifically, we first provide an overview of the development process of datasets from in-distribution and out-of-distribution perspectives. Then, we examine the evaluation metrics employed by these datasets. Third, we propose a typology that presents the development process, similarities and differences, robustness comparison, and technical features of existing debiasing methods. Furthermore, we analyze and discuss the robustness of representative vision-and-language pre-training models on VQA. Finally, through a thorough review of the available literature and experimental analysis, we discuss the key areas for future research from various viewpoints.

As a knowledge carrier, the diagram is widely distributed in many aspects of human life, such as textbooks, architectural drawings, and documents. Different from natural images, representations of visual elements in the diagram are sparser, and similar visual representations can reflect dissimilar semantics. Thus, current methods fail to capture the visual elements with precise semantics. To address this issue, regarding the aligned visual and textual elements as pairs is the way to assign the precise semantics of textual elements to visual elements. We build the first diagram dataset named align diagram element (ADE), which includes annotations for alignment relations between visual and textual elements. And we propose a visual-textual alignment model (VTAM) including graph construction and optimal aligning phases. In the graph construction phase, the relational graphs are constructed between different elements with four relational operators. The relational operators are designed to measure the relations between different elements, according to distance, connection line, inclusion, and feature similarity. In the optimal aligning phase, the representation at each visual-textual pair is improved as a weighted sum of the representations on all relational graphs. Experimental results show that our VTAM achieves a significant improvement of 10.9% on mean test folds of the ADE dataset than the current best competitor. In order to explore the role of alignment relations in diagram parsing, we introduce VTAM to diagram-related tasks, such as diagram question answering (DQA). And we achieve 2.8% to 5.9% and 4.6% to 5.1% improvements on AI2D and Foodwebs after adding VTAM. Our dataset and code are released at: https://github.com/ADE-dataset/ADE-dataset.

Commonsense knowledge graphs (CKGs) store massive commonsense knowledge as triples whose nodes consist of free-form texts. CKG reasoning aims to predict missing nodes in incomplete commonsense triples, which is challenging as it requires more accurate embeddings for reasoning. Compared to conventional knowledge graphs (KGs), CKGs have deficient structural information due to their sparsity and contain nodes indistinguishable due to the conceptual diversity. These issues limit the performance of previous reasoning methods, because they face difficulties obtaining precise CKG representations. To address these issues, we propose a context-aware CKG reasoning framework with path-guided explanations, named CoRPe. Firstly, CoRPe constructs context sentences based on the target commonsense triple using designed templates. The context captures reasoning paths instantiated from the first-order logic. Secondly, to improve CKG representations, CoRPe injects context semantics and employs a context-augmented tuning strategy on a pre-trained language model (PLM) via a synergistic optimization. Finally, CoRPe embeds structural information using a graph convolutional network (GCN) and associates the textual semantics for joint scoring. Extensive experiments on two CKGs show that CoRPe outperforms state-of-the-art KG and CKG reasoning baselines in terms of embedding and reasoning performance. Furthermore, the interpretability of CoRPe is reflected in the implicit logic during reasoning.

Relation extraction (RE) aims to identify the relation between two entities within a sentence, which plays a crucial role in information extraction. Traditional supervised setting on RE does not fit the actual scenario, due to the continuous emergence of new relations and the unavailability of massive labeled examples. Continual few-shot relation extraction (CFS-RE) is proposed as a potential solution to the above situation, which requires the model to learn new relations sequentially from a few examples. Apparently, CFS-RE is more challenging than previous RE, as the catastrophic forgetting of old knowledge and few-shot overfitting on a handful of examples. To this end, we propose a novel flexible-prompt framework on pretrained language model named FPrompt-PLM for CFS-RE, which includes flexible-prompt embedding, pretrained-language understanding, and nearest-prototype learning modules. Note that two pools in FPrompt-PLM, i.e., prompt and prototype pools, are continual updated and applied for prediction of all seen relations at current time-step. The former pool records the distinctive prompt embedding in each time period, and the latter records all learned relation prototypes. Besides, three progressive stages are introduced to learn FPrompt-PLM's parameters and apply this model for CFS-RE testing, which includes meta-training, continual meta-finetuning, and testing stages. And we improve the CFS-RE loss by incorporating multiple distillation losses as well as a novel prototype-diversity loss in these stages to alleviate the catastrophic forgetting and few-shot overfitting problems. Comprehensive experiments on two widely-used datasets show that FPrompt-PLM achieves significant performance improvements over the SOTA baselines.

Textbook question answering (TQA) is the task of correctly answering diagram or nondiagram (ND) questions given large multimodal contexts consisting of abundant essays and diagrams. In real-world scenarios, an explainable TQA system plays a key role in deepening humans' understanding of learned knowledge. However, there is no work to investigate how to provide explanations currently. To address this issue, we devise a novel architecture toward span-level eXplanations for TQA (XTQA). In this article, spans are the combinations of sentences within a paragraph. The key idea is to consider the entire textual context of a lesson as candidate evidence and then use our proposed coarse-to-fine grained explanation extracting (EE) algorithm to narrow down the evidence scope and extract the span-level explanations with varying lengths for answering different questions. The EE algorithm can also be integrated into other TQA methods to make them explainable and improve the TQA performance. Experimental results show that XTQA obtains the best overall explanation result [mean intersection over union (mIoU)] of 52.38% on the first 300 questions of CK12-QA test splits, demonstrating the explainability of our method (ND: 150 and diagram: 150). The results also show that XTQA achieves the best TQA performance of 36.46% and 36.95% on the aforementioned splits. We have released our code in https://github.com/dr-majie/opentqa.

Extracting relational triplets aims at detecting entity pairs and their semantic relations. Compared with pipeline models, joint models can reduce error propagation and achieve better performance. However, all of these models require large amounts of training data, therefore performing poorly on many long-tail relations in reality with insufficient data. In this article, we propose a novel end-to-end model, called TGIN, for few-shot triplet extraction. The core of TGIN is a multilayer heterogeneous graph with two types of nodes (entity node and relation node) and three types of edges (relation-entity edge, entity-entity edge, and relation-relation edge). On the one hand, this heterogeneous graph with entities and relations as nodes can intuitively extract relational triplets jointly, thereby reducing error propagation. On the other hand, it enables the triplet information of limited labeled data to interact better, thus maximizing the advantage of this information for few-shot triplet extraction. Moreover, we devise a graph aggregation and update method that utilizes translation algebraic operations to mine semantic features while retaining structure features between entities and relations, thereby improving the robustness of the TGIN in a few-shot setting. After updating the node and edge features through layers, TGIN propagates the label information from a few labeled examples to unlabeled examples, thus inferring triplets from these unlabeled examples. Extensive experiments on three reconstructed datasets demonstrate that TGIN can significantly improve the accuracy of triplet extraction by 2.34%~10.74% compared with the state-of-the-art baselines. To the best of our knowledge, we are the first to introduce a heterogeneous graph for few-shot relational triplet extraction.

Logical reasoning task involves diverse types of complex reasoning over text, based on the form of multiple-choice question answering (MCQA). Given the context, question and a set of options as the input, previous methods achieve superior performances on the full-data setting. However, the current benchmark dataset has the ideal assumption that the reasoning type distribution on the train split is close to the test split, which is inconsistent with many real application scenarios. To address it, there remain two problems to be studied: 1) how is the zero-shot capability of the models (train on seen types and test on unseen types)? and 2) how to enhance the perception of reasoning types for the models? For problem 1, we propose a new benchmark for generalized zero-shot logical reasoning, named ZsLR. It includes six splits based on the three type sampling strategies. For problem 2, a type-aware model TaCo is proposed. It utilizes the heuristic input reconstruction and builds a text graph with a global node. Incorporating graph reasoning and contrastive learning, TaCo can improve the type perception in the global representation. Extensive experiments on both the zero-shot and full-data settings prove the superiority of TaCo over the state-of-the-art (SOTA) methods. Also, we experiment and verify the generalization capability of TaCo on other logical reasoning dataset.

Event representation aims to transform individual events from a narrative event chain into a set of low-dimensional vectors to help support a series of downstream applications, e.g., similarity differentiation and missing event prediction. Traditional event representation models tend to focus on single modeling perspectives and thus are incapable of capturing physically disconnected yet semantically connected event segments. We, therefore, propose a heterogeneous event graph model (HeterEvent) to explicitly represent such event segments. Furthermore, another challenge in traditional event representation models is inherited from the datasets themselves. Data sparsity and insufficient labeled data are commonly encountered in event chains, easily leading to overfitting and undertraining. Therefore, we extend HeterEvent with a multistructure contrastive learning framework (MulCL) to alleviate the training risks from two structural perspectives. From the sequential perspective, a sequential-view contrastive learning component (SeqCL) is designed to facilitate the acquisition of sequential characteristics. From the graph perspective, a graph-view contrastive learning component (GraCL) is proposed to enhance the robustness of graph training by comparing different corrupted graphs. Experimental results confirm that our proposed MulCL$_{[W+E]}$model outperforms state-of-the-art baselines. Specifically, compared with the previously proposed supervised model HeterEvent$_{[W+E]}$[Zhenget al.(2020)], MulCL$_{[W+E]}$shows an average improvement of 5.3% in terms of accuracy for the inference-ability-based tasks. For the representation-ability-based tasks, MulCL$_{[W+E]}$achieves an average improvement of 2.7% in terms of accuracy for the hard similarity tasks and an improvement of 4.1% in terms of the Spearman’s correlation for the transitive sentence similarity task, respectively.

Diagram object detection is the key basis of practical applications such as textbook question answering. Because the diagram mainly consists of simple lines and color blocks, its visual features are sparser than those of natural images. In addition, diagrams usually express diverse knowledge, in which there are many low-frequency object categories in diagrams. These lead to the fact that traditional data-driven detection model is not suitable for diagrams. In this work, we propose a gestalt-perception transformer model for diagram object detection, which is based on an encoder-decoder architecture. Gestalt perception contains a series of laws to explain human perception, that the human visual system tends to perceive patches in an image that are similar, close or connected without abrupt directional changes as a perceptual whole object. Inspired by these thoughts, we build a gestalt-perception graph in transformer encoder, which is composed of diagram patches as nodes and the relationships between patches as edges. This graph aims to group these patches into objects via laws of similarity, proximity, and smoothness implied in these edges, so that the meaningful objects can be effectively detected. The experimental results demonstrate that the proposed GPTR achieves the best results in the diagram object detection task. Our model also obtains comparable results over the competitors in natural image object detection.

Extrapolation reasoning on temporal knowledge graphs (TKGs) aims to forecast future facts based on past counterparts.There are two main challenges: (1) incorporating the complex information, including structural dependencies, temporal dynamics, and hidden logical rules;(2) implementing differentiable logical rule learning and reasoning for explainability.To this end, we propose an explainable extrapolation reasoning framework TEemporal logiCal grapH networkS (TECHS), which mainly contains a temporal graph encoder and a logical decoder.The former employs a graph convolutional network with temporal encoding and heterogeneous attention to embed topological structures and temporal dynamics.The latter integrates propositional reasoning and first-order reasoning by introducing a reasoning graph that iteratively expands to find the answer.A forward message-passing mechanism is also proposed to update node representations, and their propositional and first-order attention scores.Experimental results demonstrate that it outperforms state-of-the-art baselines.

Conversational question generation aims to generate questions that depend on both context and conversation history.Conventional works utilizing deep learning have shown promising results, but heavily rely on the availability of large-scale annotated conversations.In this paper, we introduce a more realistic and less explored setting, Zero-shot Conversational Question Generation (ZeroCQG), which requires no human-labeled conversations for training.To solve ZeroCQG, we propose a multi-stage knowledge transfer framework, Synthesize, Prompt and trAnsfer with pRe-Trained lAnguage model (SPARTA) to effectively leverage knowledge from single-turn question generation instances.To validate the zero-shot performance of SPARTA, we conduct extensive experiments on three conversational datasets: CoQA, QuAC, and DoQA by transferring knowledge from three single-turn datasets: MS MARCO, NewsQA, and SQuAD.The experimental results demonstrate the superior performance of our method.Specifically, SPARTA has achieved 14.81 BLEU-4 (88.2% absolute improvement compared to T5) in CoQA with knowledge transferred from SQuAD.

Real-time bus travel time is crucial for the smart public transportation system and is beneficial for improving user satisfaction for online map services. However, it faces great challenges due to fine-grained spatial dependencies and dynamic temporal dependencies. To address the above problem, we propose GBTTE, a novel end-to-end graph attention network framework to estimate bus travel time. Specifically, we construct a novel graph structure of bus routes and use a graph attention network to capture the fine-grained spatial features of bus routes. Then, we fully exploit the joint spatial-temporal relations of bus stops through a spatial-temporal graph attention network and also capture the dynamic correlation between the route and the bus transportation network with a cross graph attention network. Finally, we integrate the route representation, the spatial-temporal representation and contextual information to estimate bus travel time. Extensive experiments carried out on two large-scale real-world datasets demonstrate the effectiveness of GBTTE. In addition, GBTTE has been deployed in production at Baidu Maps, handling tens of millions of requests every day.

Network embedding algorithms learn a mapping from the discrete representation of nodes to continuous vector spaces that preserve node proximity. Despite recent efforts to design novel models, little attention has been given to understanding the instability of network embedding. In this paper, we define the stability of node embeddings as the invariance of the nearest neighbors of nodes in different instantiations. We find that existing embedding approaches have significant amounts of instability. In addition, network structures and algorithm models influence the stability of node embeddings significantly. We also examine the implications of embedding instability for downstream tasks and find remarkable impacts on performance.

Diagram visual grounding aims to capture the correlation between language expression and local objects in the diagram, and plays an important role in the applications like textbook question answering and cross-modal retrieval. Most diagrams consist of several colors and simple geometries. This results in sparse low-level visual features, which further aggravates the gap between low-level visual and high-level semantic features of diagrams. The phenomenon brings challenges to the diagram visual grounding. To solve the above issues, we propose a gestalt-perceptual attention model to align the diagram objects and language expressions. For low-level visual features, inspired by the gestalt that simulates human visual system, we build a gestalt-perception graph network to make up the features learned by the traditional backbone network. For high-level semantic features, we design a multi-modal context attention mechanism to facilitate the interaction between diagrams and language expressions, so as to enhance the semantics of diagrams. Finally, guided by diagram features and linguistic embedding, the target query is gradually decoded to generate the coordinates of the referred object. By conducting comprehensive experiments on diagrams and natural images, we demonstrate that the proposed model achieves superior performance over the competitors. Our code will be released at https://github.com/AIProCode/GPA.

Graph contrastive learning (GCL), aiming for an embedding space where semantically similar nodes are closer, has been widely applied in graph-structured data. Researchers have proposed many approaches to define positive and negative pairs (i.e., semantically similar and dissimilar pairs) on the graph, serving as labels to learn their embedding distances. Despite the effectiveness, those approaches usually suffer from two typical learning challenges. First, the number of candidate negative pairs is enormous. Thus, it is non-trivial to select representative ones to train the model in a more effective way. Second, the heuristics (e.g., graph views or meta-path patterns) to define positive and negative pairs are sometimes less reliable, causing considerable noise for both "labelled'' positive and negative pairs. In this work, we propose a novel sampling approach B2-Sampling to address the above challenges in a unified way. On the one hand, we use balanced sampling to select the most representative negative pairs regarding both the topological and embedding diversities. On the other hand, we use biased sampling to learn and correct the labels of the most error-prone negative pairs during the training. The balanced and biased samplings can be applied iteratively for discriminating and correcting training pairs, boosting the performance of GCL models. B2-Sampling is designed as a framework to support many known GCL models. Our extensive experiments on node classification, node clustering, and graph classification tasks show that B2-Sampling significantly improves the performance of GCL models with acceptable runtime overhead. Our website[11] https://sites.google.com/view/b2-sampling/home provides access to our codes and additional experiment results.

Recently, a large amount of work has emerged for knowledge graph completion (KGC), which aims to reason over known facts and to infer the missing links. Meanwhile, contrastive learning has been applied to the KGC tasks, which can improve the representation quality of entities and relations. However, existing KGC approaches tend to improve their performance with high-dimensional embeddings and complex models, which make them suffer from large storage space and high training costs. Furthermore, contrastive loss with single positive sample learns little structural and semantic information in knowledge graphs due to the complex relation types. To address these challenges, we propose a novel knowledge graph completion model named ConKGC with the embedding dimension scaling and a relation-aware multi-positive contrastive loss. In order to achieve both space consumption reduction and model performance improvement, a new scoring function is proposed to map the raw low-dimensional embeddings of entities and relations to high-dimensional embedding space, and predict low-dimensional tail entities with latent semantic information of high-dimensional embeddings. In addition, ConKGC designs a multiple weak positive samples based contrastive loss under different relation types to maintain two important training targets, Alignment and Uniformity. This loss function and few parameters of the model ensure that ConKGC performs best and has fast convergence speed. Extensive experiments on three standard datasets confirm the effectiveness of our innovations, and the performance of ConKGC is significantly improved compared to the state-of-the-art methods.

An integral part of video analysis and surveillance is temporal activity detection, which means to simultaneously recognize and localize activities in long untrimmed videos. Currently, the most effective methods of temporal activity detection are based on deep learning, and they typically perform very well with large scale annotated videos for training. However, these methods are limited in real applications due to the unavailable videos about certain activity classes and the time-consuming data annotation. To solve this challenging problem, we propose a novel task setting called zero-shot temporal activity detection (ZSTAD), where activities that have never been seen in training still need to be detected. We design an end-to-end deep transferable network TN-ZSTAD as the architecture for this solution. On the one hand, this network utilizes an activity graph transformer to predict a set of activity instances that appear in the video, rather than produces many activity proposals in advance. On the other hand, this network captures the common semantics of seen and unseen activities from their corresponding label embeddings, and it is optimized with an innovative loss function that considers the classification property on seen activities and the transfer property on unseen activities together. Experiments on the THUMOS'14, Charades, and ActivityNet datasets show promising performance in terms of detecting unseen activities.

Textbook Question Answering (TQA) task requires answering questions by reasoning based on both the given diagrams and text context. There are mainly two challenges for the task. First, the diagrams are different from the natural images. Similar shapes or color blocks may express different semantics and there is also a large intra-topic variation for diagrams. Hence, the characteristics of visual semantic ambiguity and variable visual appearance make the diagram understanding more challenging. Second, for the text, the specific education domain with terminologies exists a great gap with the general domain. Therefore, it is difficult to represent the text semantics effectively using a text encoder pretrained in the general domain. In this paper, we propose a Spatial-Semantic Collaborative Graph Network (SSCGN) for TQA task, which can help enhance the diagram and text understanding and facilitate multimodal reasoning. Specifically, the Spatial-guided Semantic Enhancing (SSE) module fully exploits the spatial and semantic relationships between visual objects and OCR tokens to collaboratively enhance the diagram semantic understanding. Moreover, based on the semantically enhanced region representations of the SSE module, the Fine-grained Spatial-Aware Graph Network (FSA-GN) can help obtain richer relation-aware region representations for joint reasoning by capturing more fine-grained spatial relationships. We further propose multiple self-supervised auxiliary tasks to enhance the initial diagram and text semantic representations by pretraining the diagram encoder and text encoder. Extensive experiments and ablation studies are conducted to validate the effectiveness of SSCGN.

Few-shot semantic segmentation (FSS) has been developed to perform pixel-level segmentation with only a few dense labeled examples for training, which relieves the expensive annotation problem in traditional segmentation models. Current researches on FSS generally act the labeled masks on the corresponding support images to obtain the class-specific embeddings, and predict the pixel-level masks for query images by matching their pixels to these class-specific embeddings. Their performance is difficult to further break through because of the limited supervision from single-view support images and the neglect of position information from similar pixels between query and support images. To solve these issues, we propose a novel robust prior mask guided model named RPMG-FSS for the challenging FSS task. The core of RPMG-FSS is to produce a robust prior mask with good generalization ability on novel classes to better assist the following query mask prediction. Note that each element in the prior mask corresponds to one pixel in query image. It not only considers the interaction within one view and between multiple views of the support image, but also fuses the top-$k$similarity values to all support pixels and these pixels’ position information. The parameters in RPMG-FSS are optimized with the combination of segmentation loss and multi-view contrastive loss. Comprehensive experiments on two datasets show that our RPMG-FSS achieves outstanding performance comparing with the current popular baselines. The code is released onhttps://github.com/dxzxy12138/RPMG-FSS/tree/master

Diagram Question Answering (DQA) aims to correctly answer questions about given diagrams, which demands an interplay of good diagram understanding and effective reasoning. However, the same appearance of objects in diagrams can express different semantics. This kind of visual semantic ambiguity problem makes it challenging to represent diagrams sufficiently for better understanding. Moreover, since there are questions about diagrams from different perspectives, it is also crucial to perform flexible and adaptive reasoning on content-rich diagrams. In this paper, we propose a Disentangled Adaptive Visual Reasoning Network for DQA, named DisAVR, to jointly optimize the dual-process of representation and reasoning. DisAVR mainly comprises three modules: improved region feature learning, question parsing, and disentangled adaptive reasoning. Specifically, the improved region feature learning module is designed to first learn robust diagram representation by integrating detail-aware patch features and semantically-explicit text features with region features. Subsequently, the question parsing module decomposes the question into three types of question guidance including region, spatial relation and semantic relation guidance to dynamically guide subsequent reasoning. Next, the disentangled adaptive reasoning module decomposes the whole reasoning process by employing three visual reasoning cells to construct a soft fully-connected multi-layer stacked routing space. These three cells in each layer reason over object regions, semantic and spatial relations in the diagram under the corresponding question guidance. Moreover, an adaptive routing mechanism is designed to flexibly explore more optimal reasoning paths for specific diagram-question pairs. Extensive experiments on three DQA datasets demonstrate the superiority of our DisAVR.

Currently, the non-transparent computing process of deep learning has become a significant reason hindering its further development. The Neural-Symbolic (NS) system formed by integrating logic rules into neural networks has attracted increasing attention owing to its direct interpretability. Embedding symbolic logical formulas into a low-dimensional continuous space provides an effective way for the NS system. However, current studies are all constrained by the modeling ability for its syntactic structure and fail to preserve the intrinsic semantics in embeddings, which causes poor performance on downstream reasoning tasks. To this end, this paper proposes a novel method ofContrastiveGraphRepresentations (ConGR) for logical formulas embedding. First, to improve the modeling ability for the syntactic structure, ConGR introduces a densely connected graph convolutional network (GCN) with an attention mechanism to process syntax parsing graphs of formulas. In this way, discriminative local and global embeddings of formulas are obtained at the syntax level. Second, the contrastive instances (positive or negative) for each anchor formula are generated by the transformation under the guidance of logical properties. To preserve semantic information, two types of contrast, global-local and global-global, are carried out to refine formula embeddings. Extensive experiments demonstrate that ConGR obtains superior performance against state-of-the-art baselines on entailment checking and premise selection datasets.

Few-shot learning (FSL) that acquires new knowledge with little supervision, attracts much attention due to expensive cost of data annotation. Various meta-learning methods have made a great progress for few-shot problem in image and text data. In reality, data samples are not independent but rich in link relations. Large amounts of data exists in the form of graph structure such as citation, social, and biological networks. However, FSL study on graph data is still in its infancy because of the obstacle on extracting meta-knowledge from a meta node classification task. Current research just simply combines the FSL methods experienced in computer vision with node representation models together, but ignores the effect of rich links among support and query nodes in few-shot meta-task. For this issue, we propose a novel Multi-Level Graph Relation Network (MuL-GRN) for the challenging few-shot node classification. MuL-GRN extracts node embeddings through the popular graph neural networks (GNNs). And it includes a relation learning module to mine the deep node relations from three views, namely node-level, global subgraph-level, and local subgraph-level relations. For any two nodes, the node-level relation is computed on their node embeddings, global subgraph-level relation is measured on their subgraph embeddings, and the local subgraph-level relation is mined according to the pairwise node comparison information in their subgraphs. The three-view relation vectors are fused together with an interesting relation fusion module, which measures the importance of relation vector for the current few-shot classification task automatically. Extensive experiments on five real datasets show that MuL-GRN significantly outperforms existing state-of-the-art methods by a large margin.

Hierarchical context modeling plays an important role in the response generation for multi-turn conversational systems. Previous methods mainly model context as multiple independent utterances and rely on attention mechanisms to obtain the context representation. They tend to ignore the explicit responds-to relationships between adjacent utterances and the special role that the user's latest utterance (the query) plays in determining the success of a conversation. To deal with this, we propose a multi-turn response generation model named KS-CQ, which contains two crucial components, the Keep and the Select modules, to produce a neighbor-aware context representation and a context-enriched query representation. The Keep module recodes each utterance of context by attentively introducing semantics from its prior and posterior neighboring utterances. The Select module treats the context as background information and selectively uses it to enrich the query representing process. Extensive experiments on two benchmark multi-turn conversation datasets demonstrate the effectiveness of our proposal compared with the state-of-the-art baselines in terms of both automatic and human evaluations.

Visual question answering (VQA) is a task that machines should provide an accurate natural language answer given an image and a question about the image. Many studies have found that the current VQA methods are heavily driven by the surface correlation or statistical bias in the training data, and lack sufficient image grounding. To address this issue, we devise a novel end-to-end architecture that uses multitask learning to promote more sufficient image grounding and learn effective multimodality representations. The tasks consist of VQA and our proposed image cloze (IC) task requires machines to fill in the blanks accurately given an image and a textual description of the image. To ensure our model performs sufficient image grounding as much as possible, we propose a novel word-masking algorithm to develop the multimodal IC task based on the part-of-speech of words. Our model predicts the VQA answer and fills in the blanks after the multimodality representation learning that is shared by the two tasks. Experimental results show that our model achieves almost the equivalent, state-of-the-art, second-best performance on the VQA v2.0, VQA-changing priors (CP) v2, and grounded question answering (GQA) datasets, respectively, with fewer parameters and without additional data compared with baselines.

Textbook question answering (TQA) is a task that one should answer non-diagram and diagram questions accurately, given a large context which consists of abundant diagrams and essays. Although lots of studies have made significant progress in the natural image question answering (QA), they are not applicable to comprehending diagrams and reasoning over the long multimodal context. To address the above issues, we propose a relation-aware fine-grained reasoning (RAFR) network that performs fine-grained reasoning over the nodes of relation-based diagram graphs. Our method uses semantic dependencies and relative positions between nodes in the diagram to construct relation graphs and applies graph attention networks to learn diagram representations. To extract and reason over the multimodal knowledge, we first extract the text that is the most relevant to questions, options, and the instructional diagram which is the most relevant to question diagrams at the word-sentence level and the node-diagram level, respectively. Then, we apply instructional-diagram-guided attention and question-guided attention to reason over the node of question diagrams, respectively. The experimental results show that our proposed method achieves the best performance on the TQA dataset compared with baselines. We also conduct extensive ablation studies to comprehensively analyze the proposed method.

Recent research has highlighted the importance of mixed-initiative interactions in conversational search. To enable mixed-initiative interactions, information retrieval systems should be able to ask diverse questions, such as information-seeking, clarification, and open-ended ones. question generation (QG) of open-domain conversational systems aims at enhancing the interactiveness and persistence of human-machine interactions. The task is challenging because of the sparsity of question generation (QG)-specific data in conversations. Current work is limited to single-turn interaction scenarios. We propose a context-enhanced neural question generation (CNQG) model that leverages the conversational context to predict question content and pattern, then perform question decoding. A hierarchical encoder framework is employed to obtain the discourse-level context representation. Based on this, we propose Review and Transit mechanisms to respectively select contextual keywords and predict new topic words to further construct the question content. Conversational context and the predicted question content are used to produce the question pattern, which in turn guides the question decoding process implemented by a recurrent decoder with a joint attention mechanism. To fully utilize the limited QG-specific data to train our question generator, we perform multi-task learning with three auxiliary training objectives, i.e., question pattern prediction, Review , and Transit mechanisms. The required additional labeled data is obtained in a self-supervised way. We also design a weight decaying strategy to adjust the influences of various auxiliary learning tasks. To the best of our acknowledge, we are the first to extend the application of QG to the multi-turn open-domain conversational scenario. Extensive experimental results demonstrate the effectiveness of our proposal and its main components on generating relevant and informative questions, with robust performance for contexts with various lengths.

Relation prediction in knowledge graphs (KGs) aims at predicting missing relations in incomplete triples, whereas the dominant embedding paradigm has a restriction on handling unseen entities during testing.In the realworld scenario, the inductive setting is more common because entities in the training process are finite.Previous methods capture an inductive ability by implicit logic in KGs.However, it would be challenging to preciously acquire entity-independent relational semantics of compositional logic rules and to deal with the deficient supervision of logic caused by the scarcity of relational semantics.To this end, we propose a novel graph convolutional network (GCN)-based model LogCo with logical reasoning by contrastive representations.LogCo firstly extracts enclosing subgraphs and relational paths between two entities to supply the entity-independence.Then a contrastive strategy for relational path instances and the subgraph is proposed for the issue of deficient supervision.The contrastive representations are learned for a joint training regime.Finally, prediction results and logic rules for reasoning are attained.Comprehensive experiments on twelve inductive datasets show that LogCo achieves outstanding performance comparing with SOTA inductive baselines.

Relation clustering is a general approach for open relation extraction (OpenRE).Current methods have two major problems.One is that their good performance relies on large amounts of labeled and pre-defined relational instances for pre-training, which are costly to acquire in reality.The other is that they only focus on learning a high-dimensional metric space to measure the similarity of novel relations and ignore the specific relational representations of clusters.In this work, we propose a new prompt-based framework named Match-Prompt, which can realize OpenRE with efficient knowledge transfer from only a few predefined relational instances as well as mine the specific meanings for cluster interpretability.To our best knowledge, we are the first to introduce a prompt-based framework for unlabeled clustering.Experimental results on different datasets show that MatchPrompt achieves the new SOTA results for OpenRE.

Relation prediction on knowledge graphs (KGs) aims to infer missing valid triples from observed ones. Although this task has been deeply studied, most previous studies are limited to the transductive setting and cannot handle emerging entities. Actually, the inductive setting is closer to real-life scenarios because it allows entities in the testing phase to be unseen during training. However, it is challenging to precisely conduct inductive relation prediction as there exists requirements of entity-independent relation modeling and discrete logical reasoning for interoperability. To this end, we propose a novel model ConGLR to incorporate context graph with logical reasoning. Firstly, the enclosing subgraph w.r.t. target head and tail entities are extracted and initialized by the double radius labeling. And then the context graph involving relational paths, relations and entities is introduced. Secondly, two graph convolutional networks (GCNs) with the information interaction of entities and relations are carried out to process the subgraph and context graph respectively. Considering the influence of different edges and target relations, we introduce edge-aware and relation-aware attention mechanisms for the subgraph GCN. Finally, by treating the relational path as rule body and target relation as rule head, we integrate neural calculating and logical reasoning to obtain inductive scores. And to focus on the specific modeling goals of each module, the stop-gradient is utilized in the information interaction between context graph and subgraph GCNs in the training process. In this way, ConGLR satisfies two inductive requirements at the same time. Extensive experiments demonstrate that ConGLR obtains outstanding performance against state-of-the-art baselines on twelve inductive dataset versions of three common KGs.

Machine reading comprehension has aroused wide concerns, since it explores the potential of model for text understanding. To further equip the machine with the reasoning capability, the challenging task of logical reasoning is proposed. Previous works on logical reasoning have proposed some strategies to extract the logical units from different aspects. However, there still remains a challenge to model the long distance dependency among the logical units. Also, it is demanding to uncover the logical structures of the text and further fuse the discrete logic to the continuous text embedding. To tackle the above issues, we propose an end-to-end model Logiformer which utilizes a two-branch graph transformer network for logical reasoning of text. Firstly, we introduce different extraction strategies to split the text into two sets of logical units, and construct the logical graph and the syntax graph respectively. The logical graph models the causal relations for the logical branch while the syntax graph captures the co-occurrence relations for the syntax branch. Secondly, to model the long distance dependency, the node sequence from each graph is fed into the fully connected graph transformer structures. The two adjacent matrices are viewed as the attention biases for the graph transformer layers, which map the discrete logical structures to the continuous text embedding space. Thirdly, a dynamic gate mechanism and a question-aware self-attention module are introduced before the answer prediction to update the features. The reasoning process provides the interpretability by employing the logical units, which are consistent with human cognition. The experimental results show the superiority of our model, which outperforms the state-of-the-art single model on two logical reasoning benchmarks.

Abstract How Deep Neural Networks (DNNs) best cope with the understanding of multimedia contents still remains an open problem, mainly due to two factors. First, conventional DNNs cannot effectively learn the representations of the images with sparse visual information. For example, the images describing knowledge concepts in textbooks. Second, existing DNNs cannot effectively capture the fine‐grained interactions between the images and text descriptions. To address these issues, we propose a deep Cross‐Media Grouping Fusion Network (CMGFN), which mainly has two distinctive properties: 1) CMGFN can effectively learn visual features from the images with sparse visual information. This is achieved by first progressively adjusting the attention of convolution filters to valuable visual regions, and then enhancing the use of key visual information in feature construction. 2) By a cross‐media grouping co‐attention mechanism, CMGFN can effectively use the interactions between visual features of different semantics and textual descriptions, to learn cross‐media features representing different fine‐grained semantics in different groups. Empirical studies demonstrate that CMGFN not only achieves state‐of‐the‐art performance on the multimedia documents containing sparse visual information, but also shows superior general applicability on other multimedia data, e.g., the multimedia fake news.

Prevailing deep methods for image recognition require massive labeled samples in each visual category for training. However, large amounts of data annotations are timeconsuming, and some uncommon categories only have rare samples available. For this issue, we focus on more challenging few-shot learning (FSL) task, where just few labeled images are used in the training stage. Existing FSL models are constructed with various convolutional neural networks (CNNs), which are trained on an auxiliary base dataset and evaluated for new fewshot predictions on a novel dataset. The performance of these models is difficult to break through because of the domain shift between base and novel datasets and the monotonous network architectures. Considering that, we propose a novel automatic attribute consistent network called Auto-ACNet to overcome the above problems. On one hand, Auto-ACNet utilizes the attribute information about base and novel categories to guide the procedure of representation learning. It introduces the consistent and non-consistent subnets to capture the common and different attributes of image pair, which helps to mitigate the domain shift problem. On the other hand, the architecture of Auto-ACNet is searched with the popular neural architecture search (NAS) technique DARTS, for obtaining a superior FSL network automatically. And the DARTS’s search space is improved by adding the position-aware module to extract the attribute characteristics better. Extensive experimental results on two datasets indicate that the proposed Auto-ACNet achieves significant improvement over the state-of-the-art competitors in this literature.

Textbook Question Answering (TQA) is the task of answering diagram and non-diagram questions given large multi-modal contexts consisting of abundant text and diagrams. Deep text understandings and effective learning of diagram semantics are important for this task due to its specificity. In this paper, we propose a Weakly Supervised learning method for TQA (WSTQ), which regards the incompletely accurate results of essential intermediate procedures for this task as supervision to develop Text Matching (TM) and Relation Detection (RD) tasks and then employs the tasks to motivate itself to learn strong text comprehension and excellent diagram semantics respectively. Specifically, we apply the result of text retrieval to build positive as well as negative text pairs. In order to learn deep text understandings, we first pre-train the text understanding module of WSTQ on TM and then fine-tune it on TQA. We build positive as well as negative relation pairs by checking whether there is any overlap between the items/regions detected from diagrams using object detection. The RD task forces our method to learn the relationships between regions, which are crucial to express the diagram semantics. We train WSTQ on RD and TQA simultaneously, i.e., multitask learning, to obtain effective diagram semantics and then improve the TQA performance. Extensive experiments are carried out on CK12-QA and AI2D to verify the effectiveness of WSTQ. Experimental results show that our method achieves significant accuracy improvements of 5.02% and 4.12% on test splits of the above datasets respectively than the current state-of-the-art baseline. We have released our code on https://github.com/dr-majie/WSTQ.

Realtime bus waiting time information is of great importance to the intelligent public transportation system and is beneficial for improving user satisfaction by online map services. While there are limited realtime bus waiting time services in a city, because of the expensive cost of sensor deployment and sophisticated traffic conditions. To address the above problem, we propose Du-Bus, a multi-source data fusion based system, which estimates the realtime bus waiting time based on approximating the realtime locations of buses without GPS sensors, by a variety of urban datasets, including historical bus trip data reported by a limited number of GPS equipped buses, transportation network data, traffic condition data, user mobility data, and temporal data. Du-Bus approximates the realtime locations of buses without GPS sensors by jointly modeling the bus timetable and the bus realtime travel time, which can be estimated by a variety of data sources. Specifically, we first propose a BiLSTM based end-to-end model for each bus route to estimate the bus departure interval and generate the corresponding departure timetable. Then, we estimate the travel time for each individual bus via a deep neural network component by incorporating the traffic conditions, geolocation, and map query information. Finally, we estimate the bus waiting time for arbitrary stations in the city by jointly modeling the estimated bus departure timetable and travel time. We evaluate our system on two real-world datasets, and the results verify the effectiveness of Du-Bus compared with historical average based and headway based methods. Since early 2019, Du-Bus has been deployed on Baidu Maps, one of the world’s largest map services, servicing over 20 major cities in China.

Diagram is a special form of visual expression for representing complex concepts, logic, and knowledge, which widely appears in educational scenes such as textbooks, blogs, and encyclopedias. Current research on diagrams preliminarily focuses on natural disciplines such as Biology and Geography, whose expressions are still similar to natural images. In this article, we construct the first novel geometric type of diagrams dataset in Computer Science field, which has more abstract expressions and complex logical relations. The dataset has exhaustive annotations of objects and relations for about 1,300 diagrams and 3,500 question-answer pairs. We introduce the tasks of diagram classification (DC) and diagram question answering (DQA) based on the new dataset, and propose the Diagram Paring Net (DPN) that focuses on analyzing the topological structure and text information of diagrams. We use DPN-based models to solve DC and DQA tasks, and compare the performances to well-known natural images classification models and visual question answering models. Our experiments show the effectiveness of the proposed DPN-based models on diagram understanding tasks, also indicate that our dataset is more complex compared to previous natural image understanding datasets. The presented dataset opens new challenges for research in diagram understanding, and the DPN method provides a novel perspective for studying such data. Our dataset can be available from https://github.com/WayneWong97/CSDia.

Network embedding algorithms learn a mapping from the discrete representation of nodes to continuous vector spaces that preserve the proximities of nodes. The techniques have a wide range of applications in various downstream tasks such as node classification, link prediction, and network alignment. Despite recent efforts to the design of novel models, little attention has been paid to understanding the instability of network embedding. In this paper, we fill this gap by investigating several state-of-the-art network embedding methods. Node embeddings form a geometric shape in the latent space. Characterizing the geometry is critical to figure out the variance of network embedding. Hence, we define two metrics to characterize the geometric properties and find that node embeddings tremble in different instantiations of an embedding space. Then, we formally define the stability of node embeddings as the invariance of the nearest neighbors of nodes. Experimental results show that existing embedding approaches have significant amounts of instability. We explore the influence factors that affect the stability of different methods and find that both network structure and algorithm models affect the stability of node embeddings significantly. Finally, we examine the implications of embedding instability for downstream tasks and find remarkable impacts on the performance.

The quality of user-generated reviews is significant for users to understand recommendation results and make online purchasing decisions correctly. However, the reliability of a review, which captures the likelihood that a review is benign, is ignored by many studies. The low reliability reviews cause a recommendation system's unsatisfying performance. Especially the fake reviews written by fraudulent users mislead the system into generating error recommendation results and explanations, which confuse customers and deprive customers of confidence in the system. In this paper, we propose a model, Reliable Recommendation with Review-level Explanations (RRRE), which detects reliable reviews and improves the performance of the explainable recommendation system as well. Recognizing the textual content of reviews, user-item interactions are valuable features for both rating prediction and reliability prediction. RRRE builds a uniform framework to predict rating scores and reliability scores simultaneously. Firstly, RRRE embeds user preferences and item profiles, which are extracted from textual and interactive features, into the representation of the review. Secondly, the supervised information of two subtasks is jointly combined. It makes the optimization of RRRE faster and better. Finally, the reviews with both high reliability scores and rating scores are given to customers as reliable explanations. To the best of our knowledge, we are the first to consider the reliability of reviews for improving explainable recommender system. And the experimental results confirm this idea and show that our model outperforms other baseline methods on Yelp and Amazon datasets.

Diagram-sentence matching is a valuable academic research because it can help learners effectively understand the diagrams with the assisted by sentences. However, there are many uncommon objects, i.e. few-shot contents in diagrams and sentences. The existing methods for image-sentence matching have great limitations when applied to diagrams. Because they focus on the high-frequency objects during training and ignore the uncommon objects. In addition, the specialty leads to the semantic non-intuition of the diagram itself. In this work, we propose a cross-modal attention graph model for the few-shot diagram-sentence matching task named Fs-DSM. Specifically, it is composed of three modules. The graph initialization module regards the region-level diagram features and word-level sentence features as the nodes of Fs-DSM, and edges are represented as similarity between nodes. The information propagation module is a key point of Fs-DSM, in which the few-shot contents are recognized by an uncommon object recognition strategy, and then the nodes are updated by a neighborhood aggregation procedure with cross-modal propagation between all visual and textual nodes, while the edges are recomputed based on the new node features. The global association module integrates the features of regions and words to represent the global diagrams and sentences. By conducting comprehensive experiments in terms of few-shot and conventional image-sentence matching, we demonstrate that Fs-DSM achieves superior performances over the competitors on the AI2D [Formula: see text] diagram dataset and two public benchmark datasets with nature images.

Text segmentation is a fundamental step in natural language processing (NLP) and information retrieval (IR) tasks. Most existing approaches do not explicitly take into account the facet information of documents for segmentation. Text segmentation and facet annotation are often addressed as separate problems, but they operate in a common input space. This article proposes FTS, which is a novel model for faceted text segmentation via multitask learning (MTL). FTS models faceted text segmentation as an MTL problem with text segmentation and facet annotation. This model employs the bidirectional long short-term memory (Bi-LSTM) network to learn the feature representation of sentences within a document. The feature representation is shared and adjusted with common parameters by MTL, which can help an optimization model to learn a better-shared and robust feature representation from text segmentation to facet annotation. Moreover, the text segmentation is modeled as a sequence tagging task using LSTM with a conditional random fields (CRFs) classification layer. Extensive experiments are conducted on five data sets from five domains: data structure, data mining, computer network, solid mechanics, and crystallography. The results indicate that the FTS model outperforms several highly cited and state-of-the-art approaches related to text segmentation and facet annotation.

An integral part of video analysis and surveillance is temporal activity detection, which means to simultaneously recognize and localize activities in long untrimmed videos. Currently, the most effective methods of temporal activity detection are based on deep learning, and they typically perform very well with large scale annotated videos for training. However, these methods are limited in real applications due to the unavailable videos about certain activity classes and the time-consuming data annotation. To solve this challenging problem, we propose a novel task setting called zero-shot temporal activity detection (ZSTAD), where activities that have never been seen in training can still be detected. We design an end-to-end deep network based on R-C3D as the architecture for this solution. The proposed network is optimized with an innovative loss function that considers the embeddings of activity labels and their super-classes while learning the common semantics of seen and unseen activities. Experiments on both the THUMOS’14 and the Charades datasets show promising performance in terms of detecting unseen activities.

The purpose of the General Data Protection Regulation (GDPR) is to provide improved privacy protection. If an app controls personal data from users, it needs to be compliant with GDPR. However, GDPR lists general rules rather than exact step-by-step guidelines about how to develop an app that fulfills the requirements. Therefore, there may exist GDPR compliance violations in existing apps, which would pose severe privacy threats to app users. In this paper, we take mobile health applications (mHealth apps) as a peephole to examine the status quo of GDPR compliance in Android apps. We first propose an automated system, named HPDROID, to bridge the semantic gap between the general rules of GDPR and the app implementations by identifying the data practices declared in the app privacy policy and the data relevant behaviors in the app code. Then, based on HPDROID, we detect three kinds of GDPR compliance violations, including the incompleteness of privacy policy, the inconsistency of data collections, and the insecurity of data transmission. We perform an empirical evaluation of 796 mHealth apps. The results reveal that 189 (23.7%) of them do not provide complete privacy policies. Moreover, 59 apps collect sensitive data through different measures, but 46 (77.9%) of them contain at least one inconsistent collection behavior. Even worse, among the 59 apps, only 8 apps try to ensure the transmission security of collected data. However, all of them contain at least one encryption or SSL misuse. Our work exposes severe privacy issues to raise awareness of privacy protection for app users and developers.

Coreference resolution aims at recognizing different forms in a document which refer to the same entity in the real world. Although many models have been proposed and achieved success, there still exist some challenges. Recent models that use recurrent neural networks to obtain mention representations ignore dependencies between spans and their proceeding distant spans, which will lead to predicted clusters that are locally consistent but globally inconsistent. In addition, these models are trained only by maximizing the marginal likelihood of gold antecedent spans from coreference clusters, which will make some gold mentions undetectable and cause unsatisfactory coreference results. To address these challenges, we propose a neural coreference resolution model. It employs mutual attention to take into account the dependencies between spans and their proceeding spans directly (use attention mechanism to capture global information between spans and their proceeding spans). And our model is trained by jointly optimizing mention clustering and imbalanced mention detection, which enables it to detect more gold mentions in a document to make more accurate coreference decisions. Experimental results on the CoNLL-2012 English dataset show that our model can detect the most gold mentions and achieve the state-of-the-art coreference performance compared with baselines.

In the real world, a limited number of labeled finely grained images per class can hardly represent the class distribution effectively. Due to the more subtle visual differences in fine-grained images than simple images with obvious objects, that is, there exist smaller interclass and larger intraclass variations. To solve these issues, we propose an end-to-end attention-based model for fine-grained few-shot image classification (AFG) with the recent episode training strategy. It is composed mainly of a feature learning module, an image reconstruction module, and a label distribution module. The feature learning module mainly devises a 3D-Attention mechanism, which considers both the spatial positions and different channel attentions of the image features, in order to learn more discriminative local features to better represent the class distribution. The image reconstruction module calculates the mappings between local features and the original images. It is constrained by a designed loss function as auxiliary supervised information, so that the learning of each local feature does not need extra annotations. The label distribution module is used to predict the label distribution of a given unlabeled sample, and we use the local features to represent the image features for classification. By conducting comprehensive experiments on Mini-ImageNet and three fine-grained data sets, we demonstrate that the proposed model achieves superior performance over the competitors.

Image-sentence matching is a challenging task for the heterogeneity-gap between different modalities. Ranking-based methods have achieved excellent performance in this task in past decades. Given an image query, these methods typically assume that the correct matched image-sentence pair must rank before all other mismatched ones. However, this assumption may be too strict and prone to the overfitting problem, especially when some sentences in a massive database are similar and confusable with one another. In this paper, we relax the traditional ranking loss and propose a novel deep multi-modal network with a top-k ranking loss to mitigate the data ambiguity problem. With this strategy, query results will not be penalized unless the index of ground truth is outside the range of top-k query results. Considering the non-smoothness and non-convexity of the initial top-k ranking loss, we exploit a tight convex upper bound to approximate the loss and then utilize the traditional back-propagation algorithm to optimize the deep multi-modal network. Finally, we apply the method on three benchmark datasets, namely, Flickr8k, Flickr30k, and MSCOCO. Empirical results on metrics R@K (K = 1, 5, 10) show that our method achieves comparable performance in comparison to state-of-the-art methods.

The rapid growth of Android malware results in a large body of approaches devoted to malware analysis by leveraging machine learning algorithms. However, the effectiveness of these approaches primarily depends on the manual feature engineering process, which is time-consuming and labor-intensive based on expert knowledge and intuition. In this paper, we propose an automatic approach that engineers informative features from a corpus of Android malware related technical blogs, which are written in a way that mirrors the human feature engineering process. However, there are two main challenges. First, it is difficult to recognize useful knowledge in the magnanimity information of thousands of blogs. To this end, we leverage natural language processing techniques to process the blogs and extract a set of sensitive behaviors that might do harmful activities to users potentially. Second, there exists a semantic gap between the extracted sensitive behaviors and the programming language. To this end, we propose two semantic matching rules to match the behaviors with concrete code snippets such that the apps can be tested experimentally. We design and implement a system called CTDroid for malware analysis, including malware detection (MD) and familial classification (FC). After the evaluation of CTDroid on a large scale of real malware and benign apps, the experimental results demonstrate that CTDroid can achieve 95.8% true positive rate with only 1% false positive rate for MD and 97.9% accuracy for FC. Furthermore, our proposed features are more informative than those of state-of-the-art approaches.

Image captioning and visual language grounding are two important tasks for image understanding, but are seldom considered together. In this paper, we propose a Progressive Attention-Guided Network (PAGNet), which simultaneously generates image captions and predicts bounding boxes for caption words. PAGNet mainly has two distinctive properties: i) It can progressively refine the predictive results of image captioning, by updating the attention map with the predicted bounding boxes. ii) It learns bounding boxes of the words using a weakly supervised strategy, which combines the frameworks of Multiple Instance Learning (MIL) and Markov Decision Process (MDP). By using the attention map generated in the captioning process, PAGNet significantly reduces the search space of the MDP. We conduct experiments on benchmark datasets to demonstrate the effectiveness of PAGNet and results show that PAGNet achieves the best performance.

The rapid growth of Android malware has posed severe security threats to smartphone users. On the basis of the familial trait of Android malware observed by previous work, the familial analysis is a promising way to help analysts better focus on the commonalities of malware samples within the same families, thus reducing the analytical workload and accelerating malware analysis. The majority of existing approaches rely on supervised learning and face three main challenges, i.e., low accuracy, low efficiency, and the lack of labeled dataset. To address these challenges, we first construct a fine-grained behavior model by abstracting the program semantics into a set of subgraphs. Then, we propose SRA, a novel feature that depicts the similarity relationships between the Structural Roles of sensitive API call nodes in subgraphs. An SRA is obtained based on graph embedding techniques and represented as a vector, thus we can effectively reduce the high complexity of graph matching. After that, instead of training a classifier with labeled samples, we construct malware link network based on SRAs and apply community detection algorithms on it to group the unlabeled samples into groups. We implement these ideas in a system called GefDroid that performs Graph embedding based familial analysis of AnDroid malware using unsupervised learning. Moreover, we conduct extensive experiments to evaluate GefDroid on three datasets with ground truth. The results show that GefDroid can achieve high agreements (0.707-0.883 in term of NMI) between the clustering results and the ground truth. Furthermore, GefDroid requires only linear run-time overhead and takes around 8.6s to analyze a sample on average, which is considerably faster than the previous work.

Semisupervised learning aims to leverage both labeled and unlabeled data to improve performance, where most of them are graph-based methods. However, the graph-based semisupervised methods are not capable for large-scale data since the computational consumption on the construction of graph Laplacian matrix is huge. On the other hand, the substantial unlabeled data in training stage of semisupervised learning could cause large uncertainties and potential threats. Therefore, it is crucial to enhance the robustness of semisupervised classification. In this paper, a novel large-scale robust semisupervised learning method is proposed in the framework of capped ℓ2,p-norm. This strategy is superior not only in computational cost because it makes the graph Laplacian matrix unnecessary, but also in robustness to outliers since the capped ℓ2,p-norm used for loss measurement. An efficient optimization algorithm is exploited to solve the nonconvex and nonsmooth challenging problem. The complexity of the proposed algorithm is analyzed and discussed in theory detailedly. Finally, extensive experiments are conducted over six benchmark data sets to demonstrate the effectiveness and superiority of the proposed method.

Given a document stream, categorizing the topic of the documents while highlighting the key information is critical to improving user reading efficiency. Based on the practical application, previous work on topic categorization which only focus on classifying each document to a given category is not enough. It is significant to promote an interpretative topic categorization to extract the characteristic words in the documents to assist user reading. Based on the instances-bag relationship between words and the document, we propose a multiple instance learning method to tackle this problem. In addition, another problem in traditional topic categorization is the deficient word representation. In this paper, we use Bi-directional Long Short Memory (Bi-LSTM) and Convolutional neural network (CNN) to capture both the local sequential feature and the global context feature to form a comprehensive word representation to solve this problem. Finally, we thus design an effective model Interpretative Topic Categorization Model (ITCM) to exploit the MIL property with deep learning to classify the documents to the predefined topics and discover the characteristic words simultaneously. We conduct two groups of experiments and prove that ITCM not only achieves convincing performance on topic categorization, but also interpreting effectively which words characterize the document category without word-level supervised learning.

Most community question answering (CQA) websites manage plenty of question-answer pairs (QAPs) through topic-based organizations, which may not satisfy users' fine-grained search demands. Facets of topics serve as a powerful tool to navigate, refine, and group the QAPs. In this work, we propose FACM, a model to annotate QAPs with facets by extending convolution neural networks (CNNs) with a matching strategy. First, phrase information is incorporated into text representation by CNNs with different kernel sizes. Then, through a matching strategy among QAPs and facet label texts (FaLTs) acquired from Wikipedia, we generate similarity matrices to deal with the facet heterogeneity. Finally, a three-channel CNN is trained for facet label assignment of QAPs. Experiments on three real-world data sets show that FACM outperforms the state-of-the-art methods.

Due to the difficulty of collecting labeled images for hundreds of thousands of visual categories, zero-shot learning, where unseen categories do not have any labeled images in training stage, has attracted more attention. In the past, many studies focused on transferring knowledge from seen to unseen categories by projecting all category labels into a semantic space. However, the label embeddings could not adequately express the semantics of categories. Furthermore, the common semantics of seen and unseen instances cannot be captured accurately because the distribution of these instances may be quite different. For these issues, we propose a novel deep semisupervised method by jointly considering the heterogeneity gap between different modalities and the correlation among unimodal instances. This method replaces the original labels with the corresponding textual descriptions to better capture the category semantics. This method also overcomes the problem of distribution difference by minimizing the maximum mean discrepancy between seen and unseen instance distributions. Extensive experimental results on two benchmark data sets, CU200-Birds and Oxford Flowers-102, indicate that our method achieves significant improvements over previous methods.

The rapid increase in the number of Android malware poses great challenges to anti-malware systems, because the sheer number of malware samples overwhelms malware analysis systems. The classification of malware samples into families, such that the common features shared by malware samples in the same family can be exploited in malware detection and inspection, is a promising approach for accelerating malware analysis. Furthermore, the selection of representative malware samples in each family can drastically decrease the number of malware to be analyzed. However, the existing classification solutions are limited because of the following reasons. First, the legitimate part of the malware may misguide the classification algorithms because the majority of Android malware are constructed by inserting malicious components into popular apps. Second, the polymorphic variants of Android malware can evade detection by employing transformation attacks. In this paper, we propose a novel approach that constructs frequent subgraphs (fregraphs) to represent the common behaviors of malware samples that belong to the same family. Moreover, we propose and develop FalDroid, a novel system that automatically classifies Android malware and selects representative malware samples in accordance with fregraphs. We apply it to 8407 malware samples from 36 families. Experimental results show that FalDroid can correctly classify 94.2% of malware samples into their families using approximately 4.6 sec per app. FalDroid can also dramatically reduce the cost of malware investigation by selecting only 8.5% to 22% representative samples that exhibit the most common malicious behavior among all samples.

Deep convolutional neural networks (CNNs) have shown superior performance on the task of single-label image classification. However, the applicability of CNNs to multi-label images still remains an open problem, mainly because of two reasons. First, each image is usually treated as an inseparable entity and represented as one instance, which mixes the visual information corresponding to different labels. Second, the correlations amongst labels are often overlooked. To address these limitations, we propose a deep multi-modal CNN for multi-instance multi-label image classification, called MMCNN-MIML. By combining CNNs with multi-instance multi-label (MIML) learning, our model represents each image as a bag of instances for image classification and inherits the merits of both CNNs and MIML. In particular, MMCNN-MIML has three main appealing properties: 1) it can automatically generate instance representations for MIML by exploiting the architecture of CNNs; 2) it takes advantage of the label correlations by grouping labels in its later layers; and 3) it incorporates the textual context of label groups to generate multi-modal instances, which are effective in discriminating visually similar objects belonging to different groups. Empirical studies on several benchmark multi-label image data sets show that MMCNN-MIML significantly outperforms the state-of-the-art baselines on multi-label image classification tasks.

With the exponential growth of smartphone adoption, malware attacks on smartphones have resulted in serious threats to users, especially those on popular platforms, such as Android. Most Android malware is generated by piggybacking malicious payloads into benign applications (apps), which are called piggybacked apps. In this paper, we propose DAPASA, an approach to detect Android piggybacked apps through sensitive subgraph analysis. Two assumptions are established to reflect the different invocation patterns of sensitive APIs in the injected malicious payloads (rider) of a piggybacked app and in its host app (carrier). With these two assumptions, DAPASA generates a sensitive subgraph (SSG) to profile the most suspicious behavior of an app. Five features are constructed from SSG to depict the invocation patterns. The five features are fed into the machine learning algorithms to detect whether the app is piggybacked or benign. DAPASA is evaluated on a large real-world data set consisting of 2551 piggybacked apps and 44 921 popular benign apps. Extensive evaluation results demonstrate that the proposed approach exhibits an impressive detection performance compared with that of three baseline approaches even with only five numeric features. Furthermore, the proposed approach can complement permission-based approaches and API-based approaches with the combination of our five features from a new perspective of the invocation structure.

The rapid growth of Android malware poses great challenges to anti-malware systems because the sheer number of malware samples overwhelm malware analysis systems. A promising approach for speeding up malware analysis is to classify malware samples into families so that the common features in malwares belonging to the same family can be exploited for malware detection and inspection. However, the accuracy of existing classification solutions is limited because of two reasons. First, since the majority of Android malware is constructed by inserting malicious components into popular apps, the malware's legitimate part may misguide the classification algorithms. Second, the polymorphic variants of Android malware could evade the detection by employing transformation attacks. In this paper, we propose a novel approach that constructs frequent subgraph (fregraph) to represent the common behaviors of malwares in the same family for familial classification of Android malware. Moreover, we propose and develop FalDroid, an automatic system for classifying Android malware according to fregraph, and apply it to 6,565 malware samples from 30 families. The experimental results show that FalDroid can correctly classify 94.5% malwares into their families using around 4.4s per app.

No abstract available.

Indexing microblogs for realtime search is challenging, because new microblogs are created at tremendous speed, and user query requests keep constantly changing. To guarantee user obtain complete query results, micro-blogging site maintains huge indices which leads to index fragmentation or extra merging overhead during realtime search. This paper proposes an efficient LogStructured index structure with Adaptive Merging Strategy (LS-AMS) for realtime search on microblogs. LS-AMS structure consists of an inverted index buffer and a sequence of dynamically adjustable index packages with exponentially increasing sizes. These index packages manage their inverted indices using adaptive merging strategy, which can reduce the merging overhead to improve query performance and can adjust the index structure based on environmental factors, such as the arrival rate of query requests and new microblogs. Experimental results show that LS-AMS can greatly improve query performance without increasing the update cost and improve the self-adaptability in dynamic environment.

The rapidly increasing RDF data in the Linked Open Data (LOD) community project is a valuable resource for obtaining domain knowledge. However, RDF data of specific topics also shows a trend of being more decentralized and fragmented, which makes it difficult and inefficient for the users to get an overview of a specific topic and retrieve the desired information. In this paper, we demonstrate a novel system called KFM, which can aggregate the distributed RDF data of a topic according to the facets of this topic. KFM provides a new way for users to obtain and explore domain knowledge in the LOD cloud.

Real-time monitoring, providing the real-time status information of servers, is indispensable for the management of distributed systems, e.g. failure detection and resource scheduling. The scalability of fine-grained monitoring faces more and more severe challenges with scaling up distributed systems. The real-time compression which suppresses remote information update to reduce continuous monitoring cost is a promising approach to address the scalability problem. In this paper, we present the Linear Compression Algorithm (LCA) which is the application of the linear filter to real-time monitoring. To our best knowledge, existing work and LCA only explores the correlations of values of each single metric at various times. We present a novel lightweight REal-time Compression Algorithm (ReCA) which employs discovery methods of the correlation among metrics to suppress remote information update in distributed monitoring. The compression algorithms mentioned above have limited compression power because they only explore either the correlations of values of each single metric at various times or that among metrics. Therefore, we propose the Mixed Compression Algorithm (MiCA) which explores both of the correlations to achieve higher compression ratio. We implement our algorithms and an existing compression algorithm denoted by CCA in a distributed monitoring system Ganglia and conduct extensive experiments. The experimental results show that LCA and ReCA have comparable compression ratios with CCA, that MiCA achieves up to 38.2%, 27% and 44.5% higher compression ratios than CCA, LCA and ReCA with negligible overhead, respectively, and that LCA, and ReCA can both increase the scalability of Ganglia about 1.5 times and MiCA can increase about 2.33 times under a mixed-load circumstance.

Discovering hyponym relations among domain-specific terms is a fundamental task in taxonomy learning and knowledge acquisition. However, the great diversity of various domain corpora and the lack of labeled training sets make this task very challenging for conventional methods that are based on text content. The hyperlink structure of Wikipedia article pages was found to contain recurring network motifs in this study, indicating the probability of a hyperlink being a hyponym hyperlink. Hence, a novel hyponym relation extraction approach based on the network motifs of Wikipedia hyperlinks was proposed. This approach automatically constructs motif-based features from the hyperlink structure of a domain; every hyperlink is mapped to a 13-dimensional feature vector based on the 13 types of three-node motifs. The approach extracts structural information from Wikipedia and heuristically creates a labeled training set. Classification models were determined from the training sets for hyponym relation extraction. Two experiments were conducted to validate our approach based on seven domain-specific datasets obtained from Wikipedia. The first experiment, which utilized manually labeled data, verified the effectiveness of the motif-based features. The second experiment, which utilized an automatically labeled training set of different domains, showed that the proposed approach performs better than the approach based on lexico-syntactic patterns and achieves comparable result to the approach based on textual features. Experimental results show the practicability and fairly good domain scalability of the proposed approach.

Automatic term extraction is an essential task in information processing and has a very important role in many fields, such as information retrieval, knowledge acquisition. However, existing methods are mostly proposed for English domain terms, so they can not fully adapt to the term extraction from Chinese scientific texts. This paper presents a/ new approach on the analysis of the characteristics of Chinese domain terms. Firstly, we introduce a new feature which we call it “max article time” to distinguish terms from non-terms. Then, we use the classification of terms and the links between different terms to obtain the maximum discrimination of this feature. Meanwhile, our method also combines with linguistic methods. Experiments conducted on two different domains for Chinese term extraction indicate our approach has significant improvement over existing techniques and also verify the relative domain independence of the approach.

By research on topic map standard, the extended topic map (ETM) is proposed as a novel model for organization and management of the massive knowledge resources in e-learning. Based on the model, an extended topic map toolkit is designed and implemented, which allows for operations as exploration, search, consistency check and etc. The ETM toolkit not only provides learners with visual navigation and search on massive e-learning resources, but also offers an efficient way for instructors to build the shareable and reusable domain knowledge. By ETM toolkit, an extended topic map with a certain scale on computer networks has been built and is currently available for students in our university.

Aiming at deficiencies of existing knowledge resources management systems, we designed a new collaborative knowledge construction system for massive knowledge resources. By collaborative knowledge building in the following three phases: acquisition of knowledge factors, generation of the local extended topic maps and integration of global extended topic map, we realized a conjunction of concept level and knowledge element level in different knowledge granularity, as well as an integration of topic map in distributed nodes. Through those methods we achieved to establish the global extended topic map of a specific domain from knowledge resources.

The key to Deep Web crawling is to submit promising keywords to query form and retrieve Deep Web content efficiently. To select keywords, existing methods make a decision based on keywords’ statistic information deriving from TF and DF in local acquired records, thus work well only in textual databases providing full text search interfaces, whereas not well in structured databases of multi-attribute or field-restricted search interfaces. This paper proposes a novel Deep Web crawling method. Keywords are encoded as a tuple by its linguistic, statistic and HTML features so that a harvest rate evaluation model can be learned from the issued keywords for the un-issued in future. The method breaks through the assumption of plain-text search made by existing methods. Experimental results show that the method outperforms the state of the art methods.

Ranking is a central problem for information retrieval systems, because the performance of an information retrieval system is mainly evaluated by the effectiveness of its ranking results. Learning to rank has received much attention in recent years due to its importance in information retrieval. This paper focuses on learning to rank in document retrieval and presents a ranking model named OrdRank that ranks documents with ordered multiple hyperplanes. Comparison of OrdRank with other state-of-the-art ranking techniques is conducted and several evaluation criteria are employed to evaluate its performance. Experimental results on the OHSUMED dataset show that OrdRank outperforms other methods, both in terms of quality of ranking results and efficiency.

In Multi-object Multimedia Applications(MMAs), de-synchronization caused by decoding and presentation delays becomes unacceptable after playing for dozens of minutes. In order to resolve the problem, this paper presents Adaptive Multi-object Synchronization Strategy (AMSS). It obtains each media object's importance factor and synchronization requirement of each media object pair via subjective test. Then, synchronization performance metrics which considers the two factors comprehensively is proposed. Accordingly, presentation time stamps of the media objects are adjusted. Experimental results show that synchronization performance of MMA is ideal and stable if AMSS is employed.

The current behavior models of Web user are difficult to describe the semantic information of behavior. Most of them are only used to improve the performance of Web sites. To resolve the issue, a hierarchy metamodel of user behavior and a method for user behavior modeling are proposed in this paper. Based on the metadata of pages, the method can build a behavior model of Web user from three layers, i.e., URL access, activity and session. The behavior models generated by the method lay the foundation for acquiring the semantic characteristics of user behaviors. The method has been tested in our Web-based distance learning system, and it has been proved effective in describing the semantics of Web user's behavior