Modern task-oriented dialogue (TOD) systems increasingly rely on large language model (LLM) agents, leveraging Retrieval-Augmented Generation (RAG) and long-context capabilities for long-term memory utilization. However, these methods prioritise semantic similarity over task intent, degrading multi-session coherence. We propose MemGuide, a two-stage intent-driven memory selection framework: (1) Intent‑Aligned Retrieval retrieves goal-consistent QA‑formatted memory units; (2) Missing‑Slot Guided Filtering reranks units by slot-completion gain via a chain‑of‑thought reasoner and fine‑tuned LLaMA‑8B filter. We also introduce the MS-TOD, the first multi-session TOD benchmark with 132 diverse personas, 956 task goals, and annotated intent-aligned memory targets. Evaluations on MS-TOD show that MemGuide boosts task success rate by 11% (88%→99%) and reduces dialogue length by 2.84 turns, and matches single‑session performance.

Multimodal Model Editing (MMED) aims to correct erroneous knowledge in multimodal models. Existing evaluation methods, adapted from textual model editing, overstate success by relying on low-similarity or random inputs, obscure overfitting. We propose a comprehensive locality evaluation framework, covering three key dimensions: random-image locality, no-image locality, and consistent-image locality, operationalized through seven distinct data types, enabling a detailed and structured analysis of multimodal edits. We introduce De-VQA, a dynamic evaluation for visual question answering, uncovering a phenomenon we term transient blindness, overfitting to edit-similar text while ignoring visuals. Token analysis shows edits disproportionately affect textual tokens. We propose locality-aware adversarial losses to balance cross-modal representations. Empirical results demonstrate that our approach consistently outperforms existing baselines, reducing transient blindness and improving locality by 17% on average.

Recent studies have explored the capabilities of large language models (LLMs) in solving knowledge-intensive mathematical reasoning problems. However, existing benchmarks predominantly involve static theorems that LLMs have encountered during pretraining, failing to assess dynamic knowledge integration. In this work, we introduce TaxReasoning, a novel benchmark designed to evaluate LLMs’ abilities in real-world tax calculation scenarios. These tasks require not only mathematical reasoning and numerical computation, but also the extraction and application of complex, frequently updated tax regulations. Through extensive experiments with state-of-the-art LLMs using diverse prompting strategies and knowledge augmentation techniques, we uncover substantial limitations in their ability to handle dynamic, knowledge-intensive questions—primarily due to missing domain-specific knowledge and ineffective retrieval. Even the best-performing models fall significantly short of human-level performance. Our analysis points to key avenues for improvement, including enhancing LLMs' reasoning capabilities, developing more effective knowledge summarization techniques, and improving retrieval strategies. TaxReasoning offers a critical testbed for advancing LLMs in dynamic knowledge-intensive domains.

With the growing adoption of Machine-Learning-As-A-Service (MLaaS), Private Inference (PI) has emerged as a promising solution to address its security concerns through cryptographic techniques. However, nonlinear operations in neural networks account for most of the computational and communication overhead in PI. Existing studies mainly focus on optimizing and reducing the number of ReLU activations in neural networks, but traditional pruning methods may mistakenly remove ReLUs that are critical to maintaining model accuracy. To accurately evaluate the importance of ReLUs in the network, we propose ReLUPruner, a method that uses Taylor expansion to quantify the impact on loss before and after ReLU replacement. Furthermore, we establish a hierarchical importance metric to guide layer-wise ReLU budget allocation and adopt a progressive pruning strategy that dynamically adjust the pruning rate of each layer according to training progress. Extensive experiments on various models and datasets show that ReLUPruner achieves a good balance between ReLU budget and model accuracy, yielding improvements of 1.89% (12.9k ReLUs, CIFAR-10), 3.62% (50k ReLUs, CIFAR-100) and 2.66% (30k ReLUs, Tiny-ImageNet) over the previous state-of-the-art.

The identification and filtration of malicious texts in social media environments represent a significant technical challenge aimed at protecting users from online violence and disinformation. This complexity stems from the diversity and innovativeness of social media texts, which include unique expressions and special sentence structures. Particularly, malicious texts in interrogative forms pose alignment challenges with traditional corpora due to existing methods’ failure to exploit the text’s deep global semantic representations. This issue is compounded by the scant research on Chinese texts, leading to inefficiencies in recognition accuracy. To mitigate these challenges, we introduce an innovative framework based on a Global Contrastive Semantic Network (GCSN), designed to enhance malicious text recognition efficiency and accuracy by deeply learning global semantic knowledge. It comprises an encoder for global semantic information modelling and a graph-matching network for semantic similarity evaluation between question pairs, enabling the accurate identification and filtering of malicious texts with complex structures. Furthermore, we introduce a semantic consistency-based data augmentation method (COMBINE), using real-world data to generate balanced positive and negative samples, enriching the dataset and enhancing the model’s ability to distinguish semantic consistency through contrastive learning. Experimental validation on two Chinese datasets demonstrates our model’s exceptional performance, affirming its applicationa value in social media malicious text recognition. Our code is available at https://github.com/Wxy13131313131/GCSN-COMBINE

Retrieval-Augmented Generation (RAG) enhances large language models (LLMs) by integrating external knowledge, enabling them to tackle knowledge-intensive tasks. However, limited research has explored how LLMs effectively leverage RAG techniques for multi-hop question answering (QA), particularly when handling knowledge with with varying degrees of familiarity. In this paper, we introduce MINTQA (Multi-hop Question Answering on New and Tail Knowledge), a benchmark designed to evaluate multi-hop QA performance on questions involving 10,479 question-answer pairs for evaluating old/new knowledge and 17,887 pairs for assessing popular/unpopular knowledge, with each question equipped with corresponding sub-questions and answers. This benchmark primarily evaluates the multi-hop reasoning ability of LLMs and their capacity to handle knowledge with varying levels of familiarity during the reasoning process. We evaluate 22 state-of-the-art LLMs using three distinct QA strategies: LLM-based parameterized knowledge QA, direct RAG-enhanced QA, and multi-hop RAG-enhanced QA. Our experiments reveal key challenges in how LLMs handle knowledge with different familiarity and offer insights into improving their multi-hop reasoning capabilities when combined with RAG techniques.

Temporal reasoning is crucial for large language models (LLMs) to understand event concurrency and complex temporal interactions in natural language.Recent approaches rely on the LLM to infer temporal relations between events and largely overlook the inherent structural nature of temporal relationships.In this work, we propose ODL-TempLLM (Ontology-Guided and Description Logic-Constrained Temporal Reasoning with LLMs), a novel paradigm for temporal reasoning with LLMs that shifts focus from internal inference to the explicit modeling of temporal structure.ODL-TempLLM leverages ontology learning to explicitly construct structured temporal knowledge, employs a symbolic reasoner to deductively reason about temporal relations and uses logic-constrained retrieval augmentation to obtain relevant facts.Experiments results evaluated across three datasets via various LLM backbones show that our method outperforms state-of-the-art methods by 2.07-31.83F1 points and 1.00-30.73EM points, exhibiting strong generalization and highlighting the potential of explicit temporal reasoning.

Temporal reasoning remains a critical challenge for large language models (LLMs), particularly when it requires encompassing relational dependencies and numerical constraints.Yet, existing benchmarks largely overlook the joint consideration of these two dimensions and primarily rely on single-task evaluation paradigms, making it difficult to assess whether correct answers reflect grounded reasoning or arise from superficial statistical recall.To address these gaps, we introduce TNR, a benchmark designed to evaluate both Temporal Numerical and Relational reasoning.We propose a bi-directional evaluation framework consisting of forward generation via Question Answering (QA) and backward verification via Fact Verification (FV).By measuring the alignment between QA and FV, we introduce a Consistency Rate to quantify the robustness of reasoning across these two directions.Experiments on a range of LLMs reveal notable discrepancies between QA and FV performance, particularly in numerical and interval-based tasks.Moreover, our bi-directional error analysis demonstrates that these inconsistencies often stem from heuristic shortcuts and statistical co-occurrences rather than grounded logical deduction, flaws that are frequently masked in standard single-task evaluations.

Haoming Xu, Ningyuan Zhao, Yunzhi Yao, Weihong Xu, Hongru Wang, Xinle Deng, Shumin Deng, Jeff Z. Pan, Huajun Chen, Ningyu Zhang. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Boyang Xue, Bin Wu, Shuofei Qiao, Sheng Wang, Rui Wang, Yiming Du, Hongru Wang, Jeff Z. Pan, Emine Yilmaz, Kam-Fai Wong, Aldo Lipani. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Sikuan Yan, Xiufeng Yang, Zuchao Huang, Ercong Nie, Zifeng Ding, Zonggen Li, Xiaowen Ma, Jinhe Bi, Kristian Kersting, Jeff Z. Pan, Hinrich Schuetze, Volker Tresp, Yunpu Ma. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Knowledge Graph Question Answering (KGQA) aims to answer natural language questions by reasoning over structured knowledge graphs (KGs). While large language models (LLMs) have advanced KGQA through their strong reasoning capabilities, existing methods continue to struggle to fully exploit both the rich knowledge encoded in KGs and the reasoning capabilities of LLMs, particularly in complex scenarios. They often assume complete KG coverage and lack mechanisms to judge when external information is needed, and their reasoning remains locally myopic, failing to maintain coherent multi-step planning, leading to reasoning failures even when relevant knowledge exists. We propose Graph-RFT, a novel two-stage reinforcement fine-tuning KGQA framework with a ''plan–KGsearch–and–Websearch–during–think'' paradigm, that enables LLMs to perform autonomous planning and adaptive retrieval scheduling across KG and web sources under incomplete knowledge conditions. Graph-RFT introduces a chain-of-thought (CoT) fine-tuning method with a customized plan–retrieval dataset activates structured reasoning and resolves the GRPO cold-start problem. It then introduces a novel plan–retrieval guided reinforcement learning process integrates explicit planning and retrieval actions with a multi-reward design, enabling coverage-aware retrieval scheduling. It employs a Cartesian-inspired planning module to decompose complex questions into ordered sub-questions, and logical expression to guide tool invocation for globally consistent multi-step reasoning. This reasoning–retrieval process is optimized with a multi-reward combining outcome and retrieval-specific signals, enabling the model to learn when and how to combine KG and web retrieval effectively. Experiments on multiple KGQA benchmarks demonstrate that Graph-RFT achieves superior performance over strong baselines, even with smaller LLM backbones, and substantially improves complex question decomposition, factual coverage, and tool coordination.

The rapid growth of open and structured RDF data on the Web has promoted the development of dataset search as an important research topic. The core function of existing systems is ad hoc dataset retrieval (AHDR) based on the metadata of datasets, which contains limited information and often suffers from quality issues. To overcome the limitations, in this article, we systematically investigate content-based AHDR to exploit the actual RDF data in datasets. We address three main tasks of content-based AHDR with novel methods for handling the large size and complex structure of RDF data to facilitate dataset retrieval, deduplication, and snippet extraction. These methods are integrated into an online and open-source prototype called Caddie . The effectiveness and practicability of its components are evaluated on a public test collection and by a user study.

Nan Hu, Jiaoyan Chen, Yike Wu, Guilin Qi, Hongru Wang, Sheng Bi, Yongrui Chen, Tongtong Wu, Jeff Z. Pan. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Multi-hop Question Answering (MHQA) adds layers of complexity to question answering, making it more challenging.When Language Models (LMs) are prompted with multiple search results, they are tasked not only with retrieving relevant information but also employing multi-hop reasoning across the information sources.Although LMs perform well on traditional question-answering tasks, the causal mask can hinder their capacity to reason across complex contexts.In this paper, we explore how LMs respond to multi-hop questions by permuting search results (retrieved documents) under various configurations.Our study reveals interesting findings as follows: 1) Encoderdecoder models, such as the ones in the Flan-T5 family, generally outperform causal decoderonly LMs in MHQA tasks, despite being significantly smaller in size; 2) altering the order of gold documents reveals distinct trends in both Flan T5 models and fine-tuned decoderonly models, with optimal performance observed when the document order aligns with the reasoning chain order; 3) enhancing causal decoder-only models with bi-directional attention by modifying the causal mask can effectively boost their end performance.In addition to the above, we conduct a thorough investigation of the distribution of LM attention weights in the context of MHQA.Our experiments reveal that attention weights tend to peak at higher values when the resulting answer is correct.We leverage this finding to heuristically improve LMs' performance on this task.Our code is publicly available at https://github. com/hwy9855/MultiHopQA-Reasoning.

The rise of Multi-modal Pre-training highlights the necessity for a unified Multi-Modal Knowledge Graph (MMKG) representation learning framework. Such a framework is essential for embedding structured knowledge into multi-modal Large Language Models effectively, alleviating issues like knowledge misconceptions and multi-modal hallucinations. In this work, we explore the efficacy of models in accurately embedding entities within MMKGs through two pivotal tasks: Multi-modal Knowledge Graph Completion (MKGC) and Multi-modal Entity Alignment (MMEA). Building on this foundation, we propose a novel SNAG method that utilizes a Transformer-based architecture equipped with modality-level noise masking to robustly integrate multi-modal entity features in KGs. By incorporating specific training objectives for both MKGC and MMEA, our approach achieves SOTA performance across a total of ten datasets, demonstrating its versatility. Moreover, SNAG can not only function as a standalone model but also enhance other existing methods, providing stable performance improvements. Code and data are available at https://github.com/zjukg/SNAG.

Attribution is crucial in question answering (QA) with Large Language Models (LLMs). SOTA question decomposition-based approaches use long form answers to generate questions for retrieving related documents. However, the generated questions are often irrelevant and incomplete, resulting in a loss of facts in retrieval. These approaches also fail to aggregate evidence snippets from different documents and paragraphs. To tackle these problems, we propose a new fact decomposition-based framework called FIDES (faithful context enhanced fact decomposition and evidence aggregation) for attributed QA. FIDES uses a contextually enhanced two-stage faithful decomposition method to decompose long form answers into sub-facts, which are then used by a retriever to retrieve related evidence snippets. If the retrieved evidence snippets conflict with the related sub-facts, such sub-facts will be revised accordingly. Finally, the evidence snippets are aggregated according to the original sentences. Extensive evaluation has been conducted with six datasets, with an additionally proposed new metric called Attrauto–P for evaluating the evidence precision. FIDES outperforms the SOTA methods by over 14% in average with GPT-3.5-turbo, Gemini and Llama 70B series.

Information alignment evaluators are vital for various NLG evaluation tasks and trustworthy LLM deployment, reducing hallucinations and enhancing user trust.Current fine-grained methods, like FactScore, verify facts individually but neglect inter-fact dependencies, enabling subtle vulnerabilities.In this work, we introduce MONTAGELIE, a challenging benchmark that constructs deceptive narratives by "montaging" truthful statements without introducing explicit hallucinations.We demonstrate that both coarse-grained LLM-based evaluators and current fine-grained frameworks are susceptible to this attack, with AUC-ROC scores falling below 65%.To enable more robust fine-grained evaluation, we propose DOVESCORE, a novel framework that jointly verifies factual accuracy and event-order consistency.By modeling inter-fact relationships, DOVESCORE outperforms existing finegrained methods by over 8%, providing a more robust solution for long-form text alignment evaluation.Our code and datasets are available at https://github.com/dannalily/DoveScore.

In response to the call for agent-based solutions that leverage the ever-increasing capabilities of the deep models' ecosystem, we introduce a comprehensive solution for selecting appropriate models and subsequently planning a set of atomic actions to satisfy the end-users' instructions. Our system, Hive, operates over sets of models and, upon receiving natural language instructions, schedules and executes, explainable plans of atomic actions. These actions can involve one or more of the available models to achieve the overall task, while respecting end-users specific constraints. Hive is able to plan complex chains of actions while guaranteeing explainability, using an LLM-based formal logic backbone empowered by PDDL operations. We introduce the MuSE benchmark in order to offer a comprehensive evaluation of the multi-modal capabilities of agent systems. Our findings show that our framework redefines the state-of-the-art for task selection, outperforming other competing systems that plan operations across multiple models while offering transparency guarantees while fully adhering to user constraints.

Multi-modal Knowledge Graphs (MMKGs) have been widely applied across various domains for knowledge representation. However, the existing MMKGs are significantly fewer than required, and their construction faces numerous challenges, particularly in ensuring the selection of high-quality, contextually relevant images for knowledge graph enrichment. To address these challenges, we present a framework for constructing MMKGs from conventional KGs. Furthermore, to generate higher-quality images that are more relevant to the context in the given knowledge graph, we designed a neighbor selection method called Visualizable Structural Neighbor Selection (VSNS). This method consists of two modules: Visualizable Neighbor Selection (VNS) and Structural Neighbor Selection (SNS). The VNS module filters relations that are difficult to visualize, while the SNS module selects neighbors that most effectively capture the structural characteristics of the entity. To evaluate the quality of the generated images, we performed qualitative and quantitative evaluations on two datasets, MKG-Y and DB15K. The experimental results indicate that using the VSNS method to select neighbors results in higher-quality images that are more relevant to the knowledge graph.

Multimodal entity linking (MEL) aims to link ambiguous mentions within multimodal contexts to corresponding entities in a multimodal knowledge base. Most existing approaches to MEL are based on representation learning or vision-and-language pre-training mechanisms for exploring the complementary effect among multiple modalities. However, these methods suffer from two limitations. On the one hand, they overlook the possibility of considering negative samples from the same modality. On the other hand, they lack mechanisms to capture bidirectional cross-modal interaction. To address these issues, we propose a Multi-level Matching network for Multimodal Entity Linking(M3EL). Specifically, M3EL is composed of three different modules: (i) a Multimodal Feature Extraction module, which extracts modality-specific representations with a multimodal encoder and introduces an intra-modal contrastive learning sub-module to obtain better discriminative embeddings based on uni-modal differences; (ii) an Intra-modal Matching Network module, which contains two levels of matching granularity: Coarse-grained Global-to-Global and Fine-grained Global-to-Local, to achieve local and global level intra-modal interaction; (iii) a Cross-modal Matching Network module, which applies bidirectional strategies, Textual-to-Visual and Visual-to-Textual matching, to implement bidirectional cross-modal interaction. Extensive experiments conducted on WikiMEL, RichpediaMEL, and WikiDiverse datasets demonstrate the outstanding performance of M3EL when compared to the state-of-the-art baselines.

Multimodal Entity Linking (MEL) aims to link ambiguous mentions within multimodal contexts to associated entities in a multimodal knowledge base. Existing approaches to MEL introduce multimodal interaction and fusion mechanisms to bridge the modality gap and enable multi-grained semantic matching. However, they do not address two important problems: (i) mention ambiguity, i.e., the lack of semantic content caused by the brevity and omission of key information in the mention's textual context; (ii) dynamic selection of modal content, i.e., to dynamically distinguish the importance of different parts of modal information. To mitigate these issues, we propose a Multi-level Mixture of Experts (MMoE) model for MEL. MMoE has four components: (i) the description-aware mention enhancement module leverages large language models to identify the WikiData descriptions that best match a mention, considering the mention's textual context; (ii) the multimodal feature extraction module adopts multimodal feature encoders to obtain textual and visual embeddings for both mentions and entities; (iii)-(iv) the intra-level mixture of experts and inter-level mixture of experts modules apply a switch mixture of experts mechanism to dynamically and adaptively select features from relevant regions of information. Extensive experiments on WikiMEL, RichpediaMEL and WikiDiverse datasets demonstrate the outstanding performance of MMoE compared to the state-of-the-art. MMoE's code is available at: https://github.com/zhiweihu1103/MEL-MMoE.

The multimodal knowledge graph reasoning (MKGR) task aims to predict the missing facts in the incomplete MKGs by leveraging auxiliary images and descriptions of entities. Existing approaches are trained with single-target objectives, which neglect the probabilistic correlations of entity labels, especially in non-target entities. Moreover, previous studies incorporate all modalities statically or adaptively, overlooking the negative impacts of irrelevant or misleading information in the incompetent modalities. To address these issues, we introduce a novel Reinforced Multimodal Distillation framework, exploiting the Dark Side of Modalities (DSoM) from two perspectives: (1) Dark knowledge from non-target entities: We propose to train a unimodal KGR model through logit distillation to mimic the multimodal soft labels provided by pre-trained multimodal teacher models. The multimodal soft labels could provide rich supervision signals with subtle correlations among both target and non-target entities from multiple perspectives. We further decouple logits into neighbor entities and non-neighbor entities to divide into two types of correlations. (2) Dark side in unhelpful modalities: To exclude the adverse effects of unhelpful modalities, we introduce a reinforced teacher combination mechanism that dynamically selects the optimal set of multimodal teachers for each triple. The agent is trained to maximize the rewards, which are only assigned to the beneficial multimodal combination strategies for the student model. Comprehensive experiments demonstrate the effectiveness of DSoM framework on 5 MKGR datasets. Codes are available at github.com/OreOZhao/DSoM.

Jie He, Yijun Yang, Wanqiu Long, Deyi Xiong, Victor Gutierrez Basulto, Jeff Z. Pan. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

Xiongtao Zhou, Jie He, Lanyu Chen, Jingyu Li, Haojing Chen, Victor Gutierrez Basulto, Jeff Z. Pan, Hanjie Chen. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

Large Language Models (LLMs) have shown remarkable capabilities in reasoning, exemplified by the success of OpenAI-o1 and DeepSeek-R1. However, integrating reasoning with external search processes remains challenging, especially for complex multi-hop questions requiring multiple retrieval steps. We propose ReSearch, a novel framework that trains LLMs to Reason with Search via reinforcement learning without using any supervised data on reasoning steps. Our approach treats search operations as integral components of the reasoning chain, where when and how to perform searches is guided by text-based thinking, and search results subsequently influence further reasoning. We train ReSearch on Qwen2.5-7B(-Instruct) and Qwen2.5-32B(-Instruct) models and conduct extensive experiments. Despite being trained on only one dataset, our models demonstrate strong generalizability across various benchmarks. Analysis reveals that ReSearch naturally elicits advanced reasoning capabilities such as reflection and self-correction during the reinforcement learning process.

Existing frameworks for evaluating long-context language models (LCLM) can be broadly categorized into real-world applications (e.g, document summarization) and synthetic tasks (e.g, needle-in-a-haystack). Despite their utility, both approaches are accompanied by certain intrinsic limitations. Real-world tasks often involve complexity that makes interpretation challenging and suffer from data contamination, whereas synthetic tasks frequently lack meaningful coherence between the target information ("needle") and its surrounding context ("haystack"), undermining their validity as proxies for realistic applications. In response to these challenges, we posit that an ideal long-context evaluation framework should be characterized by three essential features: 1) seamless context: coherent contextual integration between target information and its surrounding context; 2) controllable setting: an extensible task setup that enables controlled studies—for example, incorporating additional required abilities such as numerical reasoning; and 3) sound evaluation: avoiding LLM-as-Judge and conduct exact-match to ensure deterministic and reproducible evaluation results.This study introduces $\textbf{LongBioBench}$, a benchmark that utilizes artificially generated biographies as a controlled environment for assessing LCLMs across dimensions of $\textit{understanding}$, $\textit{reasoning}$, and $\textit{trustworthiness}$. Our experimental evaluation, which includes $\textbf{18}$ LCLMs in total, demonstrates that most models still exhibit deficiencies in semantic understanding and elementary reasoning over retrieved results and are less trustworthy as context length increases.Our further analysis indicates some design choices employed by existing synthetic benchmarks, such as contextual non-coherence, numerical needles, and the absence of distractors, rendering them vulnerable to test the model's long-context capabilities.Moreover, we also reveal that long-context continual pretraining primarily adjusts RoPE embedding to accommodate extended context lengths, which in turn yields only marginal improvements in the model’s true capabilities. To sum up, compared to previous synthetic benchmarks, LongBioBench achieves a better trade-off between mirroring authentic language tasks and maintaining controllability, and is highly interpretable and configurable.

The identification and filtration of malicious texts in social media environments represent a significant technical challenge aimed at protecting users from online violence and disinformation. This complexity stems from the diversity and innovativeness of social media texts, which include unique expressions and special sentence structures. Particularly, malicious texts in interrogative forms pose alignment challenges with traditional corpora due to existing methods' failure to exploit the text's deep global semantic representations. This issue is compounded by the scant research on Chinese texts, leading to inefficiencies in recognition accuracy. To mitigate these challenges, we introduce an innovative framework based on a Global Contrastive Semantic Network (GCSN), designed to enhance malicious text recognition efficiency and accuracy by deeply learning global semantic knowledge. It comprises an encoder for global semantic information modelling and a graph-matching network for semantic similarity evaluation between question pairs, enabling the accurate identification and filtering of malicious texts with complex structures. Furthermore, we introduce a semantic consistency-based data augmentation method (COMBINE), using real-world data to generate balanced positive and negative samples, enriching the dataset and enhancing the model's ability to distinguish semantic consistency through contrastive learning. Experimental validation on two Chinese datasets demonstrates our model's exceptional performance, affirming its applicationa value in social media malicious text recognition. Our code is available at https://github.com/Wxy13131313131/GCSN-COMBINE

Attributed Question Answering (AQA) aims to provide both a trustworthy answer and a reliable attribution report for a given question. Retrieval is a widely adopted approach, including two general paradigms: Retrieval-Then-Read (RTR) and post-hoc retrieval. Recently, Large Language Models (LLMs) have shown remarkable proficiency, prompting growing interest in AQA among researchers. However, RTR-based AQA often suffers from irrelevant knowledge and rapidly changing information, even when LLMs are adopted, while post-hoc retrievalbased AQA struggles with comprehending long-form answers with complex logic, and precisely identifying the content needing revision and preserving the original intent. To tackle these problems, this paper proposes an Atomic fact decompositionbased Retrieval and Editing (ARE) framework, which decomposes the generated long-form answers into molecular clauses and atomic facts by the instruction-tuned LLMs. Notably, the instruction-tuned LLMs are fine-tuned using a well-constructed dataset, generated from large scale Knowledge Graphs (KGs). This process involves extracting one-hop neighbors from a given set of entities and transforming the result into coherent long-form text. Subsequently, ARE leverages a search engine to retrieve evidences related to atomic facts, inputting these evidences into an LLM-based verifier to determine whether the facts require expansion for re-retrieval or editing. Furthermore, the edited facts are backtracked into the original answer, with evidence aggregated based on the relationship between molecular clauses and atomic facts. Extensive evaluations demonstrate the superior performance of our proposed method over the state-of-the-arts on several datasets, with an additionally proposed new metricAttrpfor evaluating the precision of evidence attribution.

Logical reasoning of text requires neural models to possess strong contextual comprehension and logical reasoning ability to draw conclusions from limited information. To improve the logical reasoning capabilities of pretrained language models (PLMs), existing approaches can be broadly categorized into neural architecture-based methods and large language model (LLM)-driven strategies. While neural methods struggle with fine-grained logic that fails to capture detailed semantic roles and constraints, LLM-driven approaches, despite generating multistep reasoning sequences, lack explicit inference control and suffer from error accumulation due to their implicit and stochastic nature. Some works have tried using logical expressions, like first-order logic, but these approaches often fail to handle quantifiers systematically or support clear reasoning processes. Inspired by first-order logic and generalized quantifier (GQ) theory, we propose AQE-RF, a model based on an adaptive quantifier extension and rule-filtering graph network to address this challenge. The first component constructs a fine-grained text logical graph (FTLG) and then performs GQ instantiation based on option attention. The second component performs rule-filtered deductive reasoning, using conflict scores and dynamic programming (DP) to select coherent, interpretable inference paths. Extensive experiments on the LogiQA, ReClor, and AR-LSAT datasets demonstrate the effectiveness and robustness of AQE-RF.

Although neural models have achieved remarkable performance, they still encounter doubts due to the intransparency. To this end, model prediction explanation is attracting more and more attentions. However, current methods rarely incorporate external knowledge and still suffer from three limitations: (1) Neglecting concept completeness. Merely selecting concepts may not sufficient for prediction. (2) Lacking concept fusion. Failure to merge semantically-equivalent concepts. (3) Difficult in manipulating model behavior. Lack of verification for explanation on original model. To address these issues, we propose a novel knowledge-aware neuron interpretation framework to explain model predictions for image scene classification. Specifically, for concept completeness, we present core concepts of a scene based on knowledge graph, ConceptNet, to gauge the completeness of concepts. Our method, incorporating complete concepts, effectively provides better prediction explanations compared to baselines. Furthermore, for concept fusion, we introduce a knowledge graph-based method known as Concept Filtering, which produces over 23% point gain on neuron behaviors for neuron interpretation. At last, we propose Model Manipulation, which aims to study whether the core concepts based on ConceptNet could be employed to manipulate model behavior. The results show that core concepts can effectively improve the performance of original model by over 26%.

Event relation extraction (ERE) is a critical and fundamental challenge for natural language processing. Existing work mainly focuses on directly modeling the entire document, which cannot effectively handle long-range dependencies and information redundancy. To address these issues, we propose a cluster-aware compression method for improving event relation extraction (TacoERE), which explores a compression-then-extraction paradigm. Specifically, we first introduce document clustering for modeling event dependencies. It splits the document into intra- and inter-clusters, where intra-clusters aim to enhance the relations within the same cluster, while inter-clusters attempt to model the related events at arbitrary distances. Secondly, we utilize cluster summarization to simplify and highlight important text content of clusters for mitigating information redundancy and event distance. We have conducted extensive experiments on both pre-trained language models, such as RoBERTa, and large language models, such as ChatGPT and GPT-4, on three ERE datasets, i.e., MAVEN-ERE, EventStoryLine and HiEve. Experimental results demonstrate that TacoERE is an effective method for ERE.

We present Archer, a challenging bilingual textto-SQL dataset specific to complex reasoning, including arithmetic, commonsense and hypothetical reasoning.It contains 1,042 English questions and 1,042 Chinese questions, along with 521 unique SQL queries, covering 20 English databases across 20 domains.Notably, this dataset demonstrates a significantly higher level of complexity compared to existing publicly available datasets.Our evaluation shows that Archer challenges the capabilities of current state-of-the-art models, with a high-ranked model on the Spider leaderboard achieving only 6.73% execution accuracy on Archer test set.Thus, Archer presents a significant challenge for future research in this field.Which 4-cylinder car needs the most fuel to drive 300 miles?List how many gallons it needs, and its make and model. 开300英⾥耗油最多的四缸⻋的品牌和型号分别是什么，它 需要多少加仑的油？Commonsense Knowledge: Fuel used is calculated by divding distance driven by fuel consumption.SELECT B. Make, B.Model, 1.0 * 300 / mpg AS n_gallon FROM cars_data A JOIN car_names B ON A.Id=B.MakeId WHERE cylinders="4" ORDER BY mpg ASC LIMIT 1 Commonsense ReasoningIf all cars produced by the Daimler Benz company have 4cylinders, then in all 4-cylinder cars, which one needs the most fuel to drive 300 miles?Please list how many gallons it needs, along with its make and model.假如⽣产⾃奔驰公司的⻋都是四缸，开300英⾥耗油最多的 四缸⻋的品牌和型号分别是什么，它需要多少加仑的油？ SELECT B.Make, B.Model, 1.0 * 300 / mpg AS n_gallon FROM cars_data A JOIN car_names B ON A.id=B.makeid JOIN model_list C ON B.model=C.modelJOIN car_makers D on C.maker=D.idWHERE D.fullname="Daimler Benz" or A.cylinders="4" ORDER BY mpg ASC LIMIT 1 Hypothetical Reasoning How much higher is the maximum power of a BMW car than the maximum power of a Fiat car? 宝⻢汽⻋的最⾼功率⽐⻜雅特汽⻋的最⾼功率⾼多少？ SELECT MAX(horsepower) -(SELECT MAX (horsepower) FROM cars_data A JOIN car_names B ON A.id=B.makeid WHERE B.model="fiat") AS diff FROM cars_data A JOIN car_names B ON A.id=B.makeid WHERE B.model="bmw"

Large Language Models (LLMs) can interact with the real world by connecting with versatile external APIs, resulting in better problemsolving and task automation capabilities.Previous research primarily focuses on APIs with limited arguments from a single source or overlooks the complex dependency relationship between different APIs.However, it is essential to utilize multiple APIs collaboratively from various sources (e.g., different Apps in the iPhone), especially for complex user instructions.In this paper, we introduce AppBench, the first benchmark to evaluate LLMs' ability to plan and execute multiple APIs from various sources in order to complete the user's task.Specifically, we consider two significant challenges in multiple APIs: 1) graph structures: some APIs can be executed independently while others need to be executed one by one, resulting in graph-like execution order; and 2) permission constraints: which source is authorized to execute the API call.We have experimental results on 9 distinct LLMs; e.g., GPT-4o achieves only a 2.0% success rate at the most complex instruction, revealing that the existing state-of-the-art LLMs still cannot perform well in this situation even with the help of in-context learning and finetuning.

The abstract inference capability of the Language Model plays a pivotal role in boosting its generalization and reasoning prowess in Natural Language Inference (NLI).Entailment graphs are crafted precisely for this purpose, focusing on learning entailment relations among predicates.Yet, prevailing approaches overlook the polysemy and hierarchical nature of concepts during entity conceptualization.This oversight disregards how arguments might entail differently across various concept levels, thereby missing potential entailment connections.To tackle this hurdle, we introduce the concept pyramid and propose the HiCon-EG (Hierarchical Conceptual Entailment Graph) framework, which organizes arguments hierarchically, delving into entailment relations at diverse concept levels.By learning entailment relationships at different concept levels, the model is guided to better understand concepts so as to improve its abstract inference capabilities.Our method enhances scalability and efficiency in acquiring common-sense knowledge through leveraging statistical language distribution instead of manual labeling, Experimental results show that entailment relations derived from HiCon-EG significantly bolster abstract detection tasks.

We focus on Text-to-SQL semantic parsing from the perspective of retrieval-augmented generation.Motivated by challenges related to the size of commercial database schemata and the deployability of business intelligence solutions, we propose ASTRES that dynamically retrieves input database information and uses abstract syntax trees to select few-shot examples for in-context learning.Furthermore, we investigate the extent to which an in-parallel semantic parser can be leveraged for generating approximated versions of the expected SQL queries, to support our retrieval.We take this approach to the extreme-we adapt a model consisting of less than 500M parameters, to act as an extremely efficient approximator, enhancing it with the ability to process schemata in a parallelised manner.We apply ASTRES to monolingual and crosslingual benchmarks for semantic parsing, showing improvements over state-of-the-art baselines.Comprehensive experiments highlight the contribution of modules involved in this retrieval-augmented generation setting, revealing interesting directions for future work.

Recent studies have explored the use of Large Language Models (LLMs) with Retrieval Augmented Generation (RAG) for Knowledge Graph Question Answering (KGQA).They typically require rewriting retrieved subgraphs into natural language formats comprehensible to LLMs.However, when tackling complex questions, the knowledge rewritten by existing methods may include irrelevant information, omit crucial details, or fail to align with the question's semantics.To address them, we propose a novel rewriting method CoTKR, Chain-of-Thought Enhanced Knowledge Rewriting, for generating reasoning traces and corresponding knowledge in an interleaved manner, thereby mitigating the limitations of single-step knowledge rewriting.Additionally, to bridge the preference gap between the knowledge rewriter and the question answering (QA) model, we propose a training strategy PAQAF, Preference Alignment from Question Answering Feedback, for leveraging feedback from the QA model to further optimize the knowledge rewriter.We conduct experiments using various LLMs across several KGQA benchmarks.Experimental results demonstrate that, compared with previous knowledge rewriting methods, CoTKR generates the most beneficial knowledge representation for QA models, which significantly improves the performance of LLMs in KGQA 1 .* Equal contribution.†Corresponding author. 1 Our code is available at https://github.com/wuyike2000/CoTKR.Question: what is the unit of area that the measurement system that have an electric field strength units of volt per metre have?Answer: Square kilometer, Square meter KR Methods KR Results SummaryThe International System of Units includes various measurement systems for different physical quantities such as length, area, volume, conductance, resistance, current, force, and pressure.The measurement system with an electric field strength unit of Volt per meter does not directly relate to a specific area unit within the International System of Units.CoTKR Reason: I need to know the unit of area associated with the measurement system that has an electric field strength unit of Volt per meter.Summarize: The International System of Units includes various units for different physical quantities such as length, area, volume, etc. Reason: I need to find the specific unit of area in the International System of Units that corresponds to the electric field strength unit of Volt per meter.Summarize: The International System of Units includes units like Square kilometer, Square meter, etc. for measuring area.CoTKR+PA Reason: I need to know which measurement system includes the unit for electric field strength in Volt per meter.Knowledge: The International System of Units contains the unit for electric field strength in Volt per meter.Reason: I need to identify the unit of area within the International System of Units.Knowledge: The International System of Units includes the units Square kilometer and Square meter as area units.

Identifying and understanding user intents is a pivotal task for E-Commerce.Despite its essential role in product recommendation and business user profiling analysis, intent understanding has not been consistently defined or accurately benchmarked.In this paper, we focus on predicative user intents as "how a customer uses a product", and pose intent understanding as a natural language reasoning task, independent of product ontologies.We identify two weaknesses of FolkScope, the SOTA E-Commerce Intent Knowledge Graph: categoryrigidity and property-ambiguity.They limit its ability to strongly align user intents with products having the most desirable property, and to recommend useful products across diverse categories.Following these observations, we introduce a Product Recovery Benchmark featuring a novel evaluation framework and an example dataset.We further validate the above FolkScope weaknesses on this benchmark.Our code and dataset are available at https://github.com/stayones/Usgae-Centric- Intent-Understanding.

Yijun Yang, Jie He, Pinzhen Chen, Victor Gutierrez Basulto, Jeff Pan. Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2024.

Since the advent of large language models (LLMs), the field of tool learning has remained very active in solving various tasks in practice, including but not limited to information retrieval. This half-day tutorial provides basic concepts of this field and an overview of recent advancements with several applications. In specific, we start with some foundational components and architecture of tool learning (i.e., cognitive tool and physical tool), and then we categorize existing studies in this field into tool-augmented learning and tool-oriented learning, and introduce various learning methods to empower LLMs this kind of capability. Furthermore, we provide several cases about when, what, and how to use tools in different applications. We end with some open challenges and several potential research directions for future studies. We believe this tutorial is suited for both researchers at different stages (introductory, intermediate, and advanced) and industry practitioners who are interested in LLMs and tool learning.

Offer alignment is a key step in a product knowledge graph construction pipeline. It aims to align retailer offers of the same product for better coverage of product details. With the rapid development of online shopping services, the offer alignment task is applied in ever larger datasets. This work aims to build an offer alignment system that can efficiently be used in large-scale offer data. The key components of this system include: 1) common offer encoders for encoding text offer data into representations; 2) trainable LSH partitioning module to divide similar offers into small blocks; 3) lightweight sophisticated late-interactions for efficient filtering and scoring of offer alignment candidate pairs. We evaluate the system on public WDC offer alignment dataset, as well as DBLP-Scholar and DBLP-ACM.

Knowledge graph (KG) completion aims to find out missing triples in a KG. Some tasks, such as link prediction and instance completion, have been proposed for KG completion. They are triple-level tasks with some elements in a missing triple given to predict the missing element of the triple. However, knowing some elements of the missing triple in advance is not always a realistic setting. In this paper, we propose a novel graph-level automatic KG completion task calledTriple Set Prediction (TSP)which assumes none of the elements in the missing triples is given. TSP is to predict a set of missing triples given a set of known triples. To properly and accurately evaluate this new task, we propose 4 evaluation metrics including 3 classification metrics and 1 ranking metric, considering both the partial-open-world and the closed-world assumptions. Furthermore, to tackle the huge candidate triples for prediction, we propose a novel and efficient subgraph-based method GPHT that can predict the triple set fast. To fairly compare the TSP results, we also propose two types of methods RuleTensor-TSP and KGE-TSP applying the existing rule- and embedding-based methods for TSP as baselines. During experiments, we evaluate the proposed methods on two datasets extracted from Wikidata following the relation-similarity partial-open-world assumption proposed by us, and also create a complete family data set to evaluate TSP results following the closed-world assumption. Results prove that the methods can successfully generate a set of missing triples and achieve reasonable scores on the new task, and GPHTperforms better than the baselines with significantly shorter prediction time.

Zero-shot learning (ZSL) aims to predict unseen classes whose samples have never appeared during training. One of the most effective and widely used semantic information for zero-shot image classification are attributes which are annotations for class-level visual characteristics. However, the current methods often fail to discriminate those subtle visual distinctions between images due to not only the shortage of fine-grained annotations, but also the attribute imbalance and co-occurrence. In this paper, we present a transformer-based end-to-end ZSL method named DUET, which integrates latent semantic knowledge from the pre-trained language models (PLMs) via a self-supervised multi-modal learning paradigm. Specifically, we (1) developed a cross-modal semantic grounding network to investigate the model's capability of disentangling semantic attributes from the images; (2) applied an attribute-level contrastive learning strategy to further enhance the model's discrimination on fine-grained visual characteristics against the attribute co-occurrence and imbalance; (3) proposed a multi-task learning policy for considering multi-model objectives. We find that our DUET can achieve state-of-the-art performance on three standard ZSL benchmarks and a knowledge graph equipped ZSL benchmark. Its components are effective and its predictions are interpretable.

We propose an entity-agnostic representation learning method for handling the problem of inefficient parameter storage costs brought by embedding knowledge graphs. Conventional knowledge graph embedding methods map elements in a knowledge graph, including entities and relations, into continuous vector spaces by assigning them one or multiple specific embeddings (i.e., vector representations). Thus the number of embedding parameters increases linearly as the growth of knowledge graphs. In our proposed model, Entity-Agnostic Representation Learning (EARL), we only learn the embeddings for a small set of entities and refer to them as reserved entities. To obtain the embeddings for the full set of entities, we encode their distinguishable information from their connected relations, k-nearest reserved entities, and multi-hop neighbors. We learn universal and entity-agnostic encoders for transforming distinguishable information into entity embeddings. This approach allows our proposed EARL to have a static, efficient, and lower parameter count than conventional knowledge graph embedding methods. Experimental results show that EARL uses fewer parameters and performs better on link prediction tasks than baselines, reflecting its parameter efficiency.

Hyper-relational knowledge graphs (HKGs) extend standard knowledge graphs by associating attribute-value qualifiers to triples, which effectively represent additional fine-grained information about its associated triple. Hyper-relational knowledge graph completion (HKGC) aims at inferring unknown triples while considering its qualifiers. Most existing approaches to HKGC exploit a global-level graph structure to encode hyper-relational knowledge into the graph convolution message passing process. However, the addition of multi-hop information might bring noise into the triple prediction process. To address this problem, we propose HyperFormer, a model that considers local-level sequential information, which encodes the content of the entities, relations and qualifiers of a triple. More precisely, HyperFormer is composed of three different modules: an entity neighbor aggregator module allowing to integrate the information of the neighbors of an entity to capture different perspectives of it; a relation qualifier aggregator module to integrate hyper-relational knowledge into the corresponding relation to refine the representation of relational content; a convolution-based bidirectional interaction module based on a convolutional operation, capturing pairwise bidirectional interactions of entity-relation, entity-qualifier, and relation-qualifier. Furthermore, we introduce a Mixture-of-Experts strategy into the feed-forward layers of HyperFormer to strengthen its representation capabilities while reducing the amount of model parameters and computation. Extensive experiments on three well-known datasets with four different conditions demonstrate HyperFormer's effectiveness. Datasets and code are available at https://github.com/zhiweihu1103/HKGC-HyperFormer.

Argument mining seeks to extract arguments and their structure from unstructured texts.Identifying relations (such as attack, support, and neutral) between argumentative units is a challenging task because two units may be related to each other via implicit inferences.These inferences often rely on external commonsense knowledge to discover how one argumentative unit relates to another.State-of-the-art methods, however, rely on predefined knowledge graphs, and thus might not cover target pairs of argumentative units well.We introduce a new generative approach to finding inference chains that connect these pairs by making use of the Commonsense Transformer (COMET).We evaluate our approach on three datasets for both the two-label (attack/support) and three-label (attack/support/neutral) tasks.Our approach significantly outperforms the state-of-the-art, by 2-5% in F1 score, on two out of the three datasets with minor improvements on the remaining one.

Knowledge graph entity typing (KGET) aims at inferring plausible types of entities in knowledge graphs.Existing approaches to KGET focus on how to better encode the knowledge provided by the neighbors and types of an entity into its representation.However, they ignore the semantic knowledge provided by the way in which types can be clustered together.In this paper, we propose a novel method called Multi-view Contrastive Learning for knowledge graph Entity Typing (MCLET), which effectively encodes the coarse-grained knowledge provided by clusters into entity and type embeddings.MCLET is composed of three modules: i) Multi-view Generation and Encoder module, which encodes structured information from entity-type, entity-cluster and cluster-type views; ii) Cross-view Contrastive Learning module, which encourages different views to collaboratively improve view-specific representations of entities and types; iii) Entity Typing Prediction module, which integrates multi-head attention and a Mixture-of-Experts strategy to infer missing entity types.Extensive experiments show the strong performance of MCLET compared to the state-of-the-art.

Business Knowledge Graphs (KGs) are important to many enterprises today, providing factual knowledge and structured data that steer many products and make them more intelligent. Despite their promising benefits, building business KG necessitates solving prohibitive issues of deficient structure and multiple modalities. In this paper, we advance the understanding of the practical challenges related to building KG in non-trivial real-world systems. We introduce the process of building an open business knowledge graph (OpenBG) derived from a well-known enterprise, Alibaba Group. Specifically, we define a core ontology to cover various abstract products and consumption demands, with fine-grained taxonomy and multimodal facts in deployed applications. OpenBG is an open business KG of unprecedented scale: 2.6 billion triples with more than 88 million entities covering over 1 million core classes/concepts and 2,681 types of relations. We release all the open resources (OpenBG benchmarks) derived from it for the community and report experimental results of KG-centric tasks. We also run up an online competition based on OpenBG benchmarks, and has attracted thousands of teams. We further pre-train OpenBG and apply it to many KG-enhanced downstream tasks in business scenarios, demonstrating the effectiveness of billion-scale multimodal knowledge for e-commerce. All the resources with codes have been released at https://github.com/OpenBGBenchmark/OpenBG.

In knowledge graph completion (KGC), predicting triples involving emerging entities and/or relations, which are unseen when the KG embeddings are learned, has become a critical challenge. Subgraph reasoning with message passing is a promising and popular solution. Some recent methods have achieved good performance, but they (i) usually can only predict triples involving unseen entities alone, failing to address more realistic fully inductive situations with both unseen entities and unseen relations, and (ii) often conduct message passing over the entities with the relation patterns not fully utilized. In this study, we propose a new method named RMPI which uses a novel Relational Message Passing network for fully Inductive KGC. It passes messages directly between relations to make full use of the relation patterns for subgraph reasoning with new techniques on graph transformation, graph pruning, relation-aware neighborhood attention, addressing empty subgraphs, etc., and can utilize the relation semantics defined in the KG’s ontological schema. Extensive evaluation on multiple benchmarks has shown the effectiveness of RMPI’s techniques and its better performance compared with the existing methods that support fully inductive KGC. RMPI is also comparable to the state-of-the-art partially inductive KGC methods with very promising results achieved. Our codes, data and some supplementary experiment results are available at https://github.com/zjukg/RMPI.

In this paper, we release data about demographic information and outliers of communities of interest. Identified from Wiki-based sources, mainly Wikidata, the data covers 7.5k communities, e.g., members of the White House Coronavirus Task Force, and 345k subjects, e.g., Deborah Birx. We describe the statistical inference methodology adopted to mine such data. We release subject-centric and group-centric datasets in JSON format, as well as a browsing interface. Finally, we forsee three areas where this dataset can be useful: in social sciences research, it provides a resource for demographic analyses; in web-scale collaborative encyclopedias, it serves as an edit recommender to fill knowledge gaps; and in web search, it offers lists of salient statements about queried subjects for higher user engagement. The dataset can be accessed at: https://doi.org/10.5281/zenodo.7410436

Knowledge graphs (KGs) have become valuable knowledge resources in various applications, and knowledge graph embedding (KGE) methods have garnered increasing attention in recent years. However, conventional KGE methods still face challenges when it comes to handling unseen entities or relations during model testing. To address this issue, much effort has been devoted to various fields of KGs. In this paper, we use a set of general terminologies to unify these methods and refer to them collectively as Knowledge Extrapolation. We comprehensively summarize these methods, classified by our proposed taxonomy, and describe their interrelationships. Additionally, we introduce benchmarks and provide comparisons of these methods based on aspects that are not captured by the taxonomy. Finally, we suggest potential directions for future research.

Wenyu Huang, Mirella Lapata, Pavlos Vougiouklis, Nikos Papasarantopoulos, Jeff Pan. Proceedings of the 13th International Joint Conference on Natural Language Processing and the 3rd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers). 2023.

Pavlos Vougiouklis, Nikos Papasarantopoulos, Danna Zheng, David Tuckey, Chenxin Diao, Zhili Shen, Jeff Pan. Proceedings of the 13th International Joint Conference on Natural Language Processing and the 3rd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers). 2023.

Multi-modal entity alignment (MMEA) aims to discover identical entities across different knowledge graphs (KGs) whose entities are associated with relevant images. However, current MMEA algorithms rely on KG-level modality fusion strategies for multi-modal entity representation, which ignores the variations of modality preferences of different entities, thus compromising robustness against noise in modalities such as blurry images and relations. This paper introduces MEAformer, a mlti-modal entity alignment transformer approach for meta modality hybrid, which dynamically predicts the mutual correlation coefficients among modalities for more fine-grained entity-level modality fusion and alignment. Experimental results demonstrate that our model not only achieves SOTA performance in multiple training scenarios, including supervised, unsupervised, iterative, and low-resource settings, but also has a limited number of parameters, efficient runtime, and interpretability. Our code is available at https://github.com/zjukg/MEAformer.

We present part of Huawei's efforts in building a Product Knowledge Graph (PKG). We want to identify which product attributes (i.e. properties) are relevant and important in terms of shopping decisions to product categories (i.e. classes). This is particularly challenging when the attributes and their values are mined from online product catalogues, i.e. HTML pages. These web pages contain semi-structured data, which do not follow a concerted format and use diverse vocabulary to designate the same features. We propose a system for key attribute identification (KATIE) based on fine-tuning pre-trained models (e.g., DistilBERT) to predict the applicability and importance of an attribute to a category. We also propose an attribute synonyms identification module that allows us to discover synonymous attributes by considering not only their labels' similarities but also the similarity of their values sets. We have evaluated our approach to Huawei categories taxonomy and a set of internally mined attributes from web pages. KATIE guarantees promising performance results compared to the most recent baselines.

Machine learning (ML), especially deep neural networks, has achieved great success, but many of them often rely on a number of labeled samples for supervision. As sufficient labeled training data are not always ready due to, e.g., continuously emerging prediction targets and costly sample annotation in real-world applications, ML with sample shortage is now being widely investigated. Among all these studies, many prefer to utilize auxiliary information including those in the form of knowledge graph (KG) to reduce the reliance on labeled samples. In this survey, we have comprehensively reviewed over 90 articles about KG-aware research for two major sample shortage settings—zero-shot learning (ZSL) where some classes to be predicted have no labeled samples and few-shot learning (FSL) where some classes to be predicted have only a small number of labeled samples that are available. We first introduce KGs used in ZSL and FSL as well as their construction methods and then systematically categorize and summarize KG-aware ZSL and FSL methods, dividing them into different paradigms, such as the mapping-based, the data augmentation, the propagation-based, and the optimization-based. We next present different applications, including not only KG augmented prediction tasks such as image classification, question answering, text classification, and knowledge extraction but also KG completion tasks and some typical evaluation resources for each task. We eventually discuss some challenges and open problems from different perspectives.

The large volume of open data on the Web is expected to be reused and create value. Finding the right data to reuse is a non-trivial task addressed by the recent dataset search systems, which retrieve datasets relevant to a keyword query. An important component of such systems is snippet generation, extracting data from a retrieved dataset to exemplify its content and explain its relevance to the query. Snippet generation algorithms have emerged but were mainly evaluated by user studies. More efficient and reproducible evaluation methods are needed. To meet this challenge, in this article, we present a set of quality metrics for assessing the usefulness of a snippet from different perspectives, and we select and aggregate them into quality profiles for different stages of a dataset search process. Furthermore, we create a benchmark from thousands of collected real-world data needs and datasets, on which we apply the presented quality metrics and profiles to evaluate snippets generated by two existing algorithms and three adapted algorithms. The results, which are reproducible as they are automatically computed without human interaction, show the pros and cons of the tested algorithms and highlight directions for future research. The benchmark data is publicly available.

Commonsense knowledge about everyday concepts is an important asset for AI applications, such as question answering and chatbots. Recently, we have seen an increasing interest in the construction of structured commonsense knowledge bases (CSKBs). An important part of human commonsense is about properties that do not apply to concepts, yet existing CSKBs only store positive statements. Moreover, since CSKBs operate under the open-world assumption, absent statements are considered to have unknown truth rather than being invalid. This paper presents the UNCOMMONSENSE framework for materializing informative negative commonsense statements. Given a target concept, comparable concepts are identified in the CSKB, for which a local closed-world assumption is postulated. This way, positive statements about comparable concepts that are absent for the target concept become seeds for negative statement candidates. The large set of candidates is then scrutinized, pruned and ranked by informativeness. Intrinsic and extrinsic evaluations show that our method significantly outperforms the state-of-the-art. A large dataset of informative negations is released as a resource for future research.

We investigate the knowledge graph entity typing task which aims at inferring plausible entity types.In this paper, we propose a novel Transformer-based Entity Typing (TET) approach, effectively encoding the content of neighbors of an entity.More precisely, TET is composed of three different mechanisms: a local transformer allowing to infer missing types of an entity by independently encoding the information provided by each of its neighbors; a global transformer aggregating the information of all neighbors of an entity into a single long sequence to reason about more complex entity types; and a context transformer integrating neighbors content based on their contribution to the type inference through information exchange between neighbor pairs.Furthermore, TET uses information about class membership of types to semantically strengthen the representation of an entity.Experiments on two real-world datasets demonstrate the superior performance of TET compared to the state-of-the-art.

Multi-hop reasoning over real-life knowledge graphs (KGs) is a highly challenging problem as traditional subgraph matching methods are not capable to deal with noise and missing information. Recently, to address this problem a promising approach based on jointly embedding logical queries and KGs into a low-dimensional space to identify answer entities has emerged. However, existing proposals ignore critical semantic knowledge inherently available in KGs, such as type information. To leverage type information, we propose a novel type-aware model, TypE-aware Message Passing (TEMP), which enhances the entity and relation representation in queries, and simultaneously improves generalization, and deductive and inductive reasoning. Remarkably, TEMP is a plug-and-play model that can be easily incorporated into existing embedding-based models to improve their performance. Extensive experiments on three real-world datasets demonstrate TEMP’s effectiveness.

Knowledge Graph (KG) and its variant of ontology have been widely used for knowledge representation, and have shown to be quite effective in augmenting Zero-shot Learning (ZSL). However, existing ZSL methods that utilize KGs all neglect the intrinsic complexity of inter-class relationships represented in KGs. One typical feature is that a class is often related to other classes in different semantic aspects. In this paper, we focus on ontologies for augmenting ZSL, and propose to learn disentangled ontology embeddings guided by ontology properties to capture and utilize more fine-grained class relationships in different aspects. We also contribute a new ZSL framework named DOZSL, which contains two new ZSL solutions based on generative models and graph propagation models, respectively, for effectively utilizing the disentangled ontology embeddings. Extensive evaluations have been conducted on five benchmarks across zero-shot image classification (ZS-IMGC) and zero-shot KG completion (ZS-KGC). DOZSL often achieves better performance than the state-of-the-art, and its components have been verified by ablation studies and case studies. Our codes and datasets are available at https://github.com/zjukg/DOZSL.

Social relation extraction (SRE for short), which aims to infer the social relation between two people in daily life, has been demonstrated to be of great value in reality. Existing methods for SRE consider extracting social relation only from unimodal information such as text or image, ignoring the high coupling of multimodal information. Moreover, previous studies overlook the serious unbalance distribution on social relations. To address these issues, this paper proposes FL-MSRE, a few-shot learning based approach to extracting social relations from both texts and face images. Considering the lack of multimodal social relation datasets, this paper also presents three multimodal datasets annotated from four classical masterpieces and corresponding TV series. Inspired by the success of BERT, we propose a strong BERT based baseline to extract social relation from text only. FL-MSRE is empirically shown to outperform the baseline significantly. This demonstrates that using face images benefits text-based SRE. Further experiments also show that using two faces from different images achieves similar performance as from the same image. This means that FL-MSRE is suitable for a wide range of SRE applications where the faces of two people can only be collected from different images.

Xuefeng Su, Ru Li, Xiaoli Li, Jeff Z. Pan, Hu Zhang, Qinghua Chai, Xiaoqi Han. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

Zero-shot learning (ZSL) which aims at predicting classes that have never appeared during the training using external knowledge (a.k.a. side information) has been widely investigated. In this paper we present a literature review towards ZSL in the perspective of external knowledge, where we categorize the external knowledge, review their methods and compare different external knowledge. With the literature review, we further discuss and outlook the role of symbolic knowledge in addressing ZSL and other machine learning sample shortage issues.

Zero-shot Learning (ZSL), which aims to predict for those classes that have never appeared in the training data, has arisen hot research interests. The key of implementing ZSL is to leverage the prior knowledge of classes which builds the semantic relationship between classes and enables the transfer of the learned models (e.g., features) from training classes (i.e., seen classes) to unseen classes. However, the priors adopted by the existing methods are relatively limited with incomplete semantics. In this paper, we explore richer and more competitive prior knowledge to model the inter-class relationship for ZSL via ontology-based knowledge representation and semantic embedding. Meanwhile, to address the data imbalance between seen classes and unseen classes, we developed a generative ZSL framework with Generative Adversarial Networks (GANs).

Databases about general-world knowledge, so-called knowledge bases (KBs), are important in applications such as search and question answering. Traditionally, although KBs use open world assumption, popular KBs only store positive information, but withhold from taking any stance towards statements not contained in them. In this demo, we show that storing and presenting noteworthy negative statements would be important to overcome current limitations in various use cases. In particular, we introduce the Wiki neg ata portal, a platform to explore negative statements for Wikidata entities, by implementing a peer-based ranking method for inferring interesting negations in KBs. The demo is available at http://d5demos.mpi-inf.mpg.de/negation.

Aspect-based sentiment analysis (ABSA) aims to detect the targets (which are composed by continuous words), aspects and sentiment polarities in text. Published datasets from SemEval-2015 and SemEval-2016 reveal that a sentiment polarity depends on both the target and the aspect. However, most of the existing methods consider predicting sentiment polarities from either targets or aspects but not from both, thus they easily make wrong predictions on sentiment polarities. In particular, where the target is implicit, i.e., it does not appear in the given text, the methods predicting sentiment polarities from targets do not work. To tackle these limitations in ABSA, this paper proposes a novel method for target-aspect-sentiment joint detection. It relies on a pre-trained language model and can capture the dependence on both targets and aspects for sentiment prediction. Experimental results on the SemEval-2015 and SemEval-2016 restaurant datasets show that the proposed method achieves a high performance in detecting target-aspect-sentiment triples even for the implicit target cases; moreover, it even outperforms the state-of-the-art methods for those subtasks of target-aspect-sentiment detection that they are competent to.

Zero-shot learning (ZSL) is a popular research problem that aims at predicting for those classes that have never appeared in the training stage by utilizing the inter-class relationship with some side information. In this study, we propose to model the compositional and expressive semantics of class labels by an OWL (Web Ontology Language) ontology, and further develop a new ZSL framework with ontology embedding. The effectiveness has been verified by some primary experiments on animal image classification and visual question answering.

This paper proposes a validation mechanism for newly added triples in a growing knowledge graph. Given a logical theory, a knowledge graph, a text corpus, and a new triple to be validated, this mechanism computes a sorted list of explanations for the new triple to facilitate the validation of it, where an explanation, called an abductive text evidence, is a set of pairs of the form (triple, window) where appending the set of triples on the left to the knowledge graph enforces entailment of the new triple under the logical theory, while every sentence window on the right which is contained in the text corpus explains to some degree why the triple on the left is true. From the angle of practice, a special class of abductive text evidences called TEP-based abductive text evidence is proposed, which is constructed from explanation patterns seen before in the knowledge graph. Accordingly, a method for computing the complete set of TEP-based abductive text evidences is proposed. Moreover, a method for sorting abductive text evidences based on distantly supervised learning is proposed. To evaluate the proposed validation mechanism, four knowledge graphs with logical theories are constructed from the four great classical masterpieces of Chinese literature. Experimental results on these datasets demonstrate the efficiency and effectiveness of the proposed mechanism.

Commonsense knowledge about object properties, human behavior and general concepts is crucial for robust AI applications. However, automatic acquisition of this knowledge is challenging because of sparseness and bias in online sources. This paper presents Quasimodo, a methodology and tool suite for distilling commonsense properties from non-standard web sources. We devise novel ways of tapping into search-engine query logs and QA forums, and combining the resulting candidate assertions with statistical cues from encyclopedias, books and image tags in a corroboration step. Unlike prior work on commonsense knowledge bases, Quasimodo focuses on salient properties that are typically associated with certain objects or concepts. Extensive evaluations, including extrinsic use-case studies, show that Quasimodo provides better coverage than state-of-the-art baselines with comparable quality.

Transfer learning aims at building robust prediction models by transferring knowledge gained from one problem to another. In the semantic Web, learning tasks are enhanced with semantic representations. We exploit their semantics to augment transfer learning by dealing with when to transfer with semantic measurements and what to transfer with semantic embeddings. We further present a general framework that integrates the above measurements and embeddings with existing transfer learning algorithms for higher performance. It has demonstrated to be robust in two real-world applications: bus delay forecasting and air quality forecasting.

Relevance search over a knowledge graph (KG) has gained much research attention. Given a query entity in a KG, the problem is to find its most relevant entities. However, the relevance function is hidden and dynamic. Different users for different queries may consider relevance from different angles of semantics. The ambiguity in a query is more noticeable in the presence of thousands of types of entities and relations in a schema-rich KG, which has challenged the effectiveness and scalability of existing methods. To meet the challenge, our approach called RelSUE requests a user to provide a small number of answer entities as examples, and then automatically learns the most likely relevance function from these examples. Specifically, we assume the intent of a query can be characterized by a set of meta-paths at the schema level. RelSUE searches a KG for diversified significant meta-paths that best characterize the relevance of the user-provided examples to the query entity. It reduces the large search space of a schema-rich KG using distance and degree-based heuristics, and performs reasoning to deduplicate meta-paths that represent equivalent query-specific semantics. Finally, a linear model is learned to predict meta-path based relevance. Extensive experiments demonstrate that RelSUE outperforms several state-of-the-art methods.

In this article, the authors introduce the notion of ABox intensity in the context of predicting reasoner performance to improve the representativeness of ontology metrics, and they develop new metrics that focus on ABox features of OWL 2 EL ontologies. Their experiments show that taking into account the intensity through the proposed metrics contributes to overall prediction accuracy for ABox intensive ontologies.

Data stream learning has been largely studied for extracting knowledge structures from continuous and rapid data records. In the semantic Web, data is interpreted in ontologies and its ordered sequence is represented as an ontology stream. Our work exploits the semantics of such streams to tackle the problem of concept drift i.e., unexpected changes in data distribution, causing most of models to be less accurate as time passes. To this end we revisited (i) semantic inference in the context of supervised stream learning, and (ii) models with semantic embeddings. The experiments show accurate prediction with data from Dublin and Beijing.

This short paper motivates the need for incremental and serializable semantic reasoners. Two possible scenarios where semantic reasoners with these feature could be interesting are outlined, namely reasoning on mobile devices and managing dynamic knowledge.

Deductive reasoning and inductive learning are the most common approaches for deriving knowledge. In real world applications when data is dynamic and incomplete, especially those exposed by sensors, reasoning is limited by dynamics of data while learning is biased by data incompleteness. Therefore discovering consistent knowledge from incomplete and dynamic data is a challenging open problem. In our approach the semantics of data is captured through ontologies to empower learning (mining) with (Description Logics) reasoning. Consistent knowledge discovery is achieved by applying generic, significative, representative association semantic rules. The experiments have shown scalable, accurate and consistent knowledge discovery with data from Dublin.

Smog disaster is a type of air pollution event that negatively affects people's life and health. Forecasting smog disasters may largely reduce potential loss that they may cause. However, it is a great challenge since smog disasters are often caused by many complex factors. With the availability of huge amounts of data from the social web and physical sensors, covering information of air quality, meteorology, social event, human mobility, people's opinion, etc., it becomes possible to utilize such big data to forecast smog disasters. Especially, we can investigate the effect of social activities in smog disaster forecasting with the help of social web, which is ignored in traditional studies. In this paper, we propose a big data approach named B-Smog for smog disaster forecasting. It mainly has two components: 1) features extraction from multiple data sources to model the factors that indicate the appearance or disappearance of a smog disaster like traffic condition, human mobility, weather condition and air pollution transportation; 2) learning and predicting with heterogeneous features in multiple views. For the second component, we propose a prediction model based on an ensemble learning framework and artificial neural networks (ANNs), which achieves high accuracy in this application and can also be applied to other similar problems. We present the effectiveness of B-Smog through two cases studies in Beijing and Shanghai, and evaluate the accuracy of the prediction model through comparing it with some baselines. Moreover, the empirical findings of our study can also support decision making in smog disaster management.

The CURIOS project explores how digital archives for rural community heritage groups can be made more sustainable so that volunteer members can maintain a lasting digital presence. It is developing software tools to help remote rural communities to collaboratively maintain and present information about their cultural heritage. The objective is to investigate the use of semantic web/linked data technology to build a general, flexible and "future proof" software platform that could help such projects to develop digital archives and to be sustainable over time. As an interdisciplinary project we aim to synthesise a narrative that draws from both social science and computer science perspectives by critically reflecting upon the novel approach taken and the on-going results that are being produced.

Bar exams provide are a key watershed by which legal professionals demonstrate their knowledge of the law and its application. Passing the bar entitles one to practice the law in a given jurisdiction. The bar provides a excellent benchmark for the performance of legal information systems since passing the bar would arguably signal that the system has acquired key aspects of legal reason on a par with a human lawyer. The paper presents a first experiment with actual bar exam material, treating it as a form of textual entailment from the question to an answer. The Gold Standard is provided by the company which provides the bar exam. The paper reports a textual entailment student on a US bar exam, running ‘out of the box’ the Excitement Open Platform for textual entailment. The results and evaluation show that the tool can identify wrong answers (non-entailment) with a high F1 score, it performs poorly in identifying the correct answer (entailment). The results provide a baseline performance measure against which to evaluate future improvements. The reasons for the poor performance are examined, and proposals are made to augment the tool in the future.

For expressive ontology languages such as OWL 2 DL, classification is a computationally expensive task—2NEXPTIME-complete in the worst case. Hence, it is highly desirable to be able to accurately estimate classification time, especially for large and complex ontologies. Recently, machine learning techniques have been successfully applied to predicting the reasoning hardness category for a given (ontology, reasoner) pair. In this paper, we further develop predictive models to estimate actual classification time using regression techniques, with ontology metrics as features. Our large-scale experiments on 6 state-of-the-art OWL 2 DL reasoners and more than 450 significantly diverse ontologies demonstrate that the prediction models achieve high accuracy, good generalizability and statistical significance. Such prediction models have a wide range of applications. We demonstrate how they can be used to efficiently and accurately identify performance hotspots in a large and complex ontology, an otherwise very time-consuming and resource-intensive task.

Presenting entailments of axioms in a formal ontology is a non-trivial task. This position paper argues that the problem of selecting entailments for presentation to users is not adequately acknowledged or addressed in the literature or in implemented systems. We analyse the problem and consider some di↵erent approaches that can help to address the problem.

Forgetting is an important tool for reducing ontologies by eliminating some redundant concepts and roles while preserving sound and complete reasoning. Attempts have previously been made to address the problem of forgetting in relatively simple description logics (DLs), such as DL‐Lite and extended . However, the issue of forgetting for ontologies in more expressive DLs, such as and OWL DL, is largely unexplored. In particular, the problem of characterizing and computing forgetting for such logics is still open. In this paper, we first define semantic forgetting about concepts and roles in ontologies and state several important properties of forgetting in this setting. We then define the result of forgetting for concept descriptions in , state the properties of forgetting for concept descriptions, and present algorithms for computing the result of forgetting for concept descriptions. Unlike the case of DL‐Lite, the result of forgetting for an ontology does not exist in general, even for the special case of forgetting in TBoxes. This makes the problem of computing the result of forgetting in more challenging. We address this problem by defining a series of approximations to the result of forgetting for ontologies and studying their properties. Our algorithms for computing approximations can be directly implemented as a plug‐in of an ontology editor to enhance its ability of managing and reasoning in (large) ontologies.

The Semantic Web represents an evolution of the World Wide Web towards one of entities and their relationships, rather than pages and links. Such a progression makes it possible to represent, integrate, query and reason about structured online data. Recent years have witnessed tremendous growth of mobile computing, represented by the widespread adoption of smart phones and tablets. The versatility of such smart devices and the capabilities of semantic technologies form a great foundation for a ubiquitous Semantic Web that will contribute to further realising the true potential of both disciplines. In this paper, the authors argue for values provided by the ubiquitous Semantic Web using a mobile service discovery scenario. They also provide a brief overview of state-of-the-art research in this emerging area. Finally, the authors conclude with a summary of challenges and important research problems.

Recent years have seen a dramatic growth of semantic web on the data level, but unfortunately not on the schema level, which contains mostly concept hierarchies. The shortage of schemas makes the semantic web data difficult to be used in many semantic web applications, so schemas learning from semantic web data becomes an increasingly pressing issue. In this paper we propose a novel schemas learning approach -BelNet, which combines description logics (DLs) with Bayesian networks. In this way BelNet is capable to understand and capture the semantics of the data on the one hand, and to handle incompleteness during the learning procedure on the other hand. The main contributions of this work are: (i)we introduce the architecture of BelNet, and corresponding lypropose the ontology learning techniques in it, (ii) we compare the experimental results of our approach with the state-of-the-art ontology learning approaches, and provide discussions from different aspects.

Due to the increasing volume of and interconnections between semantic datasets, it becomes a challenging task for novice users to know what are included in a dataset, how they can make use of them, and particularly, what queries should be asked. In this paper we analyse several types of candidate insightful queries and propose a framework to generate such queries and identify their relations. To verify our approach, we implemented our framework and evaluated its performance with benchmark and real world datasets.

As the semantic web expands, ontological data becomes distributed over a large network of data sources on the Web. Consequently, evaluating queries that aim to tap into this distributed semantic database necessitates the ability to consult multiple data sources efficiently. In this paper, we propose methods and heuristics to efficiently query distributed ontological data based on a series of properties of summarized data. In our approach, each source summarizes its data as another RDF graph, and relevant section of these summaries are merged and analyzed at query evaluation time. We show how the analysis of these summaries enables more efficient source selection, query pruning and transformation of expensive distributed joins into local joins.

In coalition operations, information from different sources belong to different organisations have to be gathered and aggregated. The information from these resources may not be consistent. Inconsistencies in the gathered information creates severe uncertainties that hinders the usefulness of the information. In this paper, we have propose a Subjective Logic based approach for modelling the trustworthiness of information sources within a specific context. This model is used to handle inconsistencies through filtering information from less trustworthy sources.

In order to directly reason over inconsistent OWL 2 DL ontologies, this paper considers linear order inference which comes from propositional logic. Consequences of this inference in an inconsistent ontology are defined as consequences in a certain consistent sub-ontology. This paper proposes a novel framework for compiling an OWL 2 DL ontology to a Horn propositional program so that the intended consistent sub-ontology for linear order inference can be approximated from the compiled result in polynomial time. A tractable method is proposed to realize this framework. It guarantees that the compiled result has a polynomial size. Experimental results show that the proposed method computes the exact intended sub-ontology for almost all test cases, while it is significantly more efficient and scalable than state-of-the-art exact methods.

ABox abduction is an important reasoning facility in Description Logics (DLs). It finds all minimal sets of ABox axioms, called abductive solutions, which should be added to a background ontology to enforce entailment of an observation which is a specified set of ABox axioms. However, ABox abduction is far from practical by now because there lack feasible methods working in finite time for expressive DLs. To pave a way to practical ABox abduction, this paper proposes a new problem for ABox abduction and a new method for computing abductive solutions accordingly. The proposed problem guarantees finite number of abductive solutions. The proposed method works in finite time for a very expressive DL, , which underpins the W3C standard language OWL 2, and guarantees soundness and conditional completeness of computed results. Experimental results on benchmark ontologies show that the method is feasible and can scale to large ABoxes.

The vision of the Semantic Web is to build a ‘Web of data’ that enables machines to understand the semantics of information on the Web. The Linked Open Data (LOD) project encourages people and organizations to publish various open data sets as Resource Description Framework (RDF) on the Web, which promotes the development of the Semantic Web. Among various LOD datasets, DBpedia has proved a successful structured knowledge base, and has become the central interlinking-hub of the Web of data in English. However, in the Chinese language, there is little linked data published and linked to DBpedia. This hinders the structured knowledge sharing of both Chinese and cross-lingual resources. This paper deals with an approach for building a large-scale Chinese structured knowledge base from Chinese wiki resources, including Hudong and Baidu Baike. The proposed approach first builds an ontology based on the wiki category system and infoboxes, and then extracts instances from wiki articles. Using Hudong as our source, our approach builds an ontology containing 19 542 concepts and 2381 properties. 802 593 instances are extracted and described using the concepts and properties in the extracted ontology and 62 679 of them are linked to equivalent instances in DBpedia. As from Baidu Baike, our approach builds an ontology containing 299 concepts, 37 object properties, and 5590 data type properties. 1 319 703 instances are extracted from Baidu Baike, and 84 343 of them are linked to instances in DBpedia. We provide RDF dumps and SPARQL endpoint to access the established Chinese knowledge bases. The knowledge bases built using our approach can be used not only in Chinese linked data building, but also in many useful applications of large-scale knowledge bases, such as question-answering and semantic search.

The six papers in this special issue depict the state of the art and practice of the impact of semantic technologies in the field of Software Engineering.

ABox abduction is an important aspect for abductive reasoning in Description Logics (DLs). It finds all minimal sets of ABox axioms that should be added to a background ontology to enforce entailment of a specified set of ABox axioms. As far as we know, by now there is only one ABox abduction method in expressive DLs computing abductive solutions with certain minimality. However, the method targets an ABox abduction problem that may have infinitely many abductive solutions and may not output an abductive solution in finite time. Hence, in this paper we propose a new ABox abduction problem which has only finitely many abductive solutions and also propose a novel method to solve it. The method reduces the original problem to an abduction problem in logic programming and solves it with Prolog engines. Experimental results show that the method is able to compute abductive solutions in benchmark OWL DL ontologies with large ABoxes.

So far researchers in the Description Logics / Ontology communities mainly consider ontology reasoning services for static ontologies. The rapid development of the Semantic Web and its emerging data ask for reasoning technologies for dynamic knowledge streams. Existing work on stream reasoning is focused on lightweight languages such as RDF and RDFS. In this paper, we introduce the notion of Ontology Stream Management System (OSMS) and present a stream-reasoning approach based on Truth Maintenance System (TMS). We present optimised EL++ algorithm to reduce memory consumption. Our evaluations show that the optimisation improves TMS-enabled EL++ reasoning to deal with relatively large volumes of data and update efficiently.

Ontology reasoning and reasoning-based search expansion are powerful tools for semantic applications. In this paper, we investigate how to apply ontology reasoning, with different ontology metamodeling approaches, to diagnose software models, in particular on models with multiple levels software models, in particular on models with multiple levels of abstraction.

Computing all diagnoses of an inconsistent ontology is important in ontology-based applications. However, the number of diagnoses can be very large. It is impractical to enumerate all diagnoses before identifying the target one to render the ontology consistent. Hence, we propose to represent all diagnoses by multiple sets of partial diagnoses, where the total number of partial diagnoses can be small and the target diagnosis can be directly retrieved from these partial diagnoses. We also propose methods for computing the new representation of all diagnoses in an OWL DL ontology. Experimental results show that computing the new representation of all diagnoses is much easier than directly computing all diagnoses.

Possibilistic logic provides a convenient tool for dealing with uncertainty and handling inconsistency. In this paper, we propose possibilistic description logics as an extension of description logics, which are a family of well-known ontology languages. We first give the syntax and semantics of possibilistic description logics and define several inference services in possibilistic description logics. We show that these inference serviced can be reduced to the task of computing the inconsistency degree of a knowledge base in possibilistic description logics. Since possibilistic inference services suffer from the drowning problem, that is, axioms whose confidence degrees are less than or equal to the inconsistency are not used, we consider a drowning-free variant of possibilistic inference, called linear order inference. We propose an algorithm for computing the inconsistency degree of a possibilistic description logic knowledge base and an algorithm for the linear order inference. We consider the impact of our possibilistic description logics on ontology learning and ontology merging. Finally, we implement these algorithms and provide some interesting evaluation results. © 2011 Wiley Periodicals, Inc.

Large scale ontology applications require efficient and robust description logic (DL) reasoning services. Expressive DLs usually have very high worst case complexity while tractable DLs are restricted in terms of expressive power. This brings a new challenge: can users use expressive DLs to build their ontologies and still enjoy the efficient services as in tractable languages. In this paper, we present a soundness preserving approximate reasoning framework for TBox reasoning in OWL2-DL. The ontologies are encoded into EL++ with additional data structures. A tractable algorithm is presented to classify such approximation by realizing more and more inference patterns. Preliminary evaluation shows that our approach can classify existing benchmarks in large scale efficiently with a high recall.

In image retrieval, most existing approaches that incorporate local features produce high dimensional vectors, which lead to a high computational and data storage cost. Moreover, when it comes to the retrieval of generic real-life images, randomly generated patches are often more discriminant than the ones produced by corner/blob detectors. In order to tackle these problems, we propose a novel method incorporating local features with a hybrid sampling (a combination of detector-based and random sampling). We take three large data collections for the evaluation: MIRFlickr, ImageCLEF, and a collection from British National Geological Survey. The overall performance of the proposed approach is better than the performance of global features and comparable with the current state-of-the-art methods in content-based image retrieval. One of the advantages of our method when compared with others is its easy implementation and low computational cost. Another is that hybrid sampling can improve the performance of other methods based on the ``bag of visual words'' approach.

Giorgos Flouris, Mathieu d'Aquin, Grigoris Antoniou, Jeff Pan, Dimitris Plexousakis; Special Issue on Ontology Dynamics, Journal of Logic and Computation,

This paper gives a brief overview of the International Workshop on Semantic Technologies in System Maintenance. It describes a number of semantic technologies (e.g., ontologies, text mining, and knowledge integration techniques) and identifies diverse tasks in software maintenance where the use of semantic technologies can be beneficial, such as traceability, system comprehension, software artifact analysis, and information integration.

Fuzzy ontologies are envisioned to be useful in the Semantic Web. Existing fuzzy ontology reasoners are not scalable enough to handle the scale of data that the Web provides. In this paper, we propose a framework of fuzzy query languages for fuzzy ontologies, and present query answering algorithms for these query languages over fuzzy DL-Lite ontologies. Moreover, this paper reports on implementation of our approach in the fuzzy DL-Lite query engine in the ONTOSEARCH2 system and preliminary, but encouraging, benchmarking results. To the best of our knowledge, this is the first ever scalable query engine for fuzzy ontologies.

We consider continuous quantities that are used to describe the physical world, such as colour, shape, sound, texture and spatial and temporal arrangements. Natural languages are not adept at describing these quantities, nor are they easily incorporated into ontologies in the form of discrete terms. In this article, we analyse the way that natural languages handle continuous quantities, propose a general semantics based on metric spaces, and describe how to treat semantic values computationally, so that we may automate the processing of texts which describe continuous quantities allowing, for example, query evaluation and the integration of multiple texts. This provides a basis for incorporating these quantities into ontologies and combining their semantics with automated reasoning tools. We run a series of experiments to evaluate the semantics, the general framework, and the computational system we have developed.

Information integration and retrieval have been important problems for many information systems-it is hard to combine multidimensional and parallel information and make them available for application queries. In our previous work [12], we have shown how to use ontologies to facilitate integrating and querying parallel but single dimensional information. In this paper, we further investigate how to take advantage of ontologies to facilitate integrating parallel information and querying over multiple quantities.

It is widely recognized today that the management of imprecision and vagueness will yield more intelligent and realistic knowledge-based applications. Description Logics (DLs) are a family of knowledge representation languages that have gained considerable attention the last decade, mainly due to their decidability and the existence of empirically high performance of reasoning algorithms. In this paper, we extend the well known fuzzy ALC DL to the fuzzy SHIN DL, which extends the fuzzy ALC DL with transitive role axioms (S), inverse roles (I), role hierarchies (H) and number restrictions (N). We illustrate why transitive role axioms are difficult to handle in the presence of fuzzy interpretations and how to handle them properly. Then we extend these results by adding role hierarchies and finally number restrictions. The main contributions of the paper are the decidability proof of the fuzzy DL languages fuzzy-SI and fuzzy-SHIN, as well as decision procedures for the knowledge base satisfiability problem of the fuzzy-SI and fuzzy-SHIN.

Knowledge-based systems in the semantic Web era can make use of the power of the semantic Web languages and technologies, in particular those related to ontologies. Recent research has shown that user-defined data types are very useful for semantic Web and ontology applications. The W3C semantic Web best practices and development working group has set up a task force to address this issue. Very recently, OWL-Eu and OWL-E, two decidable extensions of the W3C standard ontology language OWL DL, have been proposed to support customized data types and customized data type predicates, respectively. In this paper, we propose a flexible reasoning architecture for these two expressive semantic Web ontology languages and describe our prototype implementation of the reasoning architecture, based on the well-known FaCT DL reasoner, which witnesses the two key flexibility features of our proposed architecture: 1) It allows users to define their own data types and data type predicates based on built-in ones and 2) new data type reasoners can be added into the architecture without having to change the concept reasoner

Semantic Web (SW) languages are supposed to be compatible with each other in a meaningful way, so as to facilitate machine understanding. Recent research, however, shows that the semantics of the standard SW annotation language RDF (as well as its ontological extension RDFS) and that of the standard SW ontology language OWL DL are not compatible with each other. This paper investigates some issues behind this incompatibility and proposes a novel modification of RDF(S) as a firm semantic foundation for many of the latest description logics-based SW ontology languages, including OWL DL. Furthermore, the bidirectional one-to-one mapping between RDFS(FA) axioms in strata 0-2 and OWL DL axioms has been established, which enables RDFS(FA)-agents and OWL DL-agents to communicate with each other more easily. As a result, the introduction of RDFS(FA) clarifies the vision of the semantic Web and solidifies RDF(S)'s proposed role as the base of the semantic Web

In this paper we present a proposal for representing soft constraint satisfaction problems (CSPs) within the Semantic Web architecture. The proposal is motivated by the need for a service-providing agent in a virtual organisation to reason about its commitments as soft constraints. The three essential requirements addressed are: (1) the need to have constraints express commitments in terms of Semantic Web services, (2) the need to associate utility values with constraints, to reflect the relative importance of satisfying them, and (3) the need to make statements about which constraints are satisfied and violated by a given solution. The proposal builds upon previous work in defining a Semantic Web Constraint Interchange Format (CIF), which itself builds on the proposed Semantic Web Rule Language (SWRL). The paper describes an ontology for representing soft CSPs and their solutions, allowing an agent’s set of commitments to be expressed as a collection of soft constraints. The ontology is an open interchange format for soft CSPs, allowing commitment to be communicated and exchanged among the members of a virtual organisation. 1.

In the Semantic Web, inconsistencies in OWL on- tologies may easily occur. Existing approaches ei- ther identify the minimally unsatisfiable sub-ontologies or calculate the maximally satisfiable sub-ontologies. However practical problems remain; it is not clear which axioms or which parts of axioms should be se- lected for repair, and how to repair those axioms. In this paper, we address this limitation by proposing a fine-grained approach to resolving unsatisfiable ontolo- gies. We revise the axiom tracing technique first pro- posed by Baader and Hollunder, so as to track which parts of the problematic axioms cause the unsatisfiabil- ity. Moreover, we support ontology users in rewriting problematic axioms. In order to minimise the impact of changes and prevent unintended entailment loss, harm- ful and helpful changes are identified and provided as guidelines. Based on the methods described we present a preliminary version of an interactive debugging tool and demonstrate its applicability in practice.

This paper proposes OWL FA, a decidable extension of OWL DL with the metamodeling architecture of RDFS(FA). It shows that the knowledge base satisfiability problem of OWL FA can be reduced to that of OWL DL, and compares the FA semantics with the recently proposed contextual semantics and Hilog semantics for OWL.