Recent advances in Multimodal Large Language Models (MLLMs) have significantly pushed the frontier of egocentric video question answering (EgocentricQA). However, existing benchmarks and studies are mainly limited to common daily activities such as cooking and cleaning. In contrast, real-world deployment inevitably encounters domain shifts, where target domains differ substantially in both visual style and semantic content. To bridge this gap, we introduce EgoCross, a comprehensive benchmark designed to evaluate the cross-domain generalization of MLLMs in EgocentricQA. EgoCross covers four diverse and challenging domains, including surgery, industry, extreme sports, and animal perspective, representing realistic and high-impact application scenarios. It comprises approximately 1,000 QA pairs across 798 video clips, spanning four key QA tasks: prediction, recognition, localization, and counting. Each QA pair provides both OpenQA and CloseQA formats to support fine-grained evaluation. Extensive experiments show that most existing MLLMs, whether general-purpose or egocentric-specialized, struggle to generalize to domains beyond daily life, highlighting the limitations of current models. Furthermore, we conduct several pilot studies, e.g., fine-tuning and reinforcement learning, to explore potential improvements. We hope EgoCross and our accompanying analysis will serve as a foundation for advancing domain-adaptive, robust egocentric video understanding.

The emergence of fine-tuning-as-a-service has revealed a new vulnerability in large language models (LLMs). A mere handful of malicious data uploaded by users can subtly manipulate the fine-tuning process, leading to a compromised alignment state. Existing methods to counteract fine-tuning attacks typically require substantial computational resources. Even with parameter-efficient techniques like LoRA, gradient updates remain essential. To address these challenges, we propose Neuron-Level Safety Realignment (NLSR), a training-free framework that restores the safety of LLMs based on the similarity difference of safety-critical neurons before and after fine-tuning. The core of our framework is first to construct a safety reference model from an initially aligned model to amplify safety-related features in neurons. We then utilize this reference model to identify safety-critical neurons, which we prepare as patches. Finally, we selectively restore only those neurons that exhibit significant similarity differences by transplanting these prepared patches, thereby minimally altering the fine-tuned model. Extensive experiments demonstrate significant safety enhancements in fine-tuned models across multiple downstream tasks, while greatly maintaining task-level accuracy. Our findings indicate that safety-critical neurons exhibit significant regional variations after fine-tuning, which can be effectively corrected through neuron transplantation from the reference model without the need for additional training.

In the rapidly evolving landscape of large language models (LLMs) for medical applications, ensuring the reliability and accuracy of these models in clinical settings is paramount. Existing benchmarks often focus on fixed-format tasks like multiple-choice QA, which fail to capture the complexity of real-world clinical diagnostics. Moreover, traditional evaluation metrics and LLM-based evaluators struggle with misalignment, often providing oversimplified assessments that do not adequately reflect human judgment. To address these challenges, we introduce HDCEval, a Hierarchical Divide-and-Conquer Evaluation framework tailored for fine-grained alignment in medical evaluation. HDCEval is built on a set of fine-grained medical evaluation guidelines developed in collaboration with professional doctors, encompassing Patient Question Relevance, Medical Knowledge Correctness, and Expression. The framework decomposes complex evaluation tasks into specialized subtasks, each evaluated by expert models trained through Attribute-Driven Token Optimization (ADTO) on a meticulously curated preference dataset. This hierarchical approach ensures that each aspect of the evaluation is handled with expert precision, leading to a significant improvement in alignment with human evaluators.

Xinyu Wang, Changzhi Sun, Lian Cheng, Yuanbin Wu, Dell Zhang, Xiaoling Wang, Xuelong Li. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

With the proliferation of large language models (LLMs) in the medical domain, there is increasing demand for improved evaluation techniques to assess their capabilities. However, traditional metrics like F1 and ROUGE, which rely on token overlaps to measure quality, significantly overlook the importance of medical terminology. While human evaluation tends to be more reliable, it can be very costly and may as well suffer from inaccuracies due to limits in human expertise and motivation. Although there are some evaluation methods based on LLMs, their usability in the medical field is limited due to their proprietary nature or lack of expertise. To tackle these challenges, we present AutoMedEval, an open-sourced automatic evaluation model with 13B parameters specifically engineered to measure the question-answering proficiency of medical LLMs. The overarching objective of AutoMedEval is to assess the quality of responses produced by diverse models, aspiring to significantly reduce the dependence on human evaluation. Specifically, we propose a hierarchical training method involving curriculum instruction tuning and an iterative knowledge introspection mechanism, enabling AutoMedEval to acquire professional medical assessment capabilities with limited instructional data. Human evaluations indicate that AutoMedEval surpasses other baselines in terms of correlation with human judgments.

As multimodal large language models (MLLMs) gain prominence in the medical field, the need for precise evaluation methods to assess their effectiveness has become critical. While benchmarks provide a reliable means to evaluate the capabilities of MLLMs, traditional metrics like ROUGE and BLEU employed for open domain evaluation only focus on token overlap and may not align with human judgment. While human evaluation is more reliable, it is labor-intensive, costly, and not scalable. LLM-based evaluation methods have proven promising, but to date, there is still an urgent need for open-source multimodal LLM-based evaluators in the medical field. To address this issue, we introduce ACE-M^3, an open-sourced Automatic Capability Evaluator for Multimodal Medical Models that specifically designed to assess the question answering abilities of medical MLLMs. It first utilizes a branch-merge architecture to provide both detailed analysis and a concise final score based on standard medical evaluation criteria. Subsequently, a reward token-based direct preference optimization (RTDPO) strategy is incorporated to save training time without compromising performance of our model. Extensive experiments have demonstrated the effectiveness of our ACE-M^3 model in evaluating the capabilities of medical MLLMs.

Semi-Supervised change detection (SSCD) aims to detect changes between bi-temporal remote sensing images by utilizing limited labeled data and abundant unlabeled data. Existing methods struggle in complex scenarios, exhibiting poor performance when confronted with noisy data. They typically neglect intra-layer multi-scale features while emphasizing inter-layer fusion, harming the integrity of change objects with different scales. In this paper, we propose HSACNet, a Hierarchical Scale-Aware Consistency regularized Network for SSCD. Specifically, we integrate Segment Anything Model 2 (SAM2), using its Hiera backbone as the encoder to extract inter-layer multi-scale features and applying adapters for parameter-efficient fine-tuning. Moreover, we design a Scale-Aware Differential Attention Module (SADAM) that can precisely capture intra-layer multi-scale change features and suppress noise. Additionally, a dual-augmentation consistency regularization strategy is adopted to effectively utilize the unlabeled data. Extensive experiments across four CD benchmarks demonstrate that our HSACNet achieves state-of-the-art performance, with reduced parameters and computational cost.

Traditional Medicine (TM) is the oldest healthcare form and has been increasingly adopted as the primary or complementary medical therapy in the world. However, TM's practical development remains highly challenging. While artificial intelligence (AI) has become powerful in advancing modern medicine, limited attention has been paid to its potential and usage in TM. This study addresses this gap through a probe-based interview study with 16 TM clinicians, examining their experiences, perceptions, and expectations of AI-empowered clinical support systems. Our findings reveal that despite numerous AI-CDS systems, their practical usage in TM settings was still limited. We identify a series of practical challenges when integrating AI-CDS into TM clinical scenarios, largely due to TM's unique features and the significant data work challenges these features present. We end by critically discussing the potential issues that may arise when integrating AI into practical TM scenarios, and proposing a series of practical recommendations for future studies.

Prescription recommendation is critical for clinical decision support in Traditional Chinese Medicine (TCM), aiming to recommend a herb set based on a patient's symptoms. The core principle of TCM clinical practice, treatment based on syndrome differentiation (SD), follows a four-step progressive process: symptoms to syndromes, therapeutic methods, and herbs. However, existing models oversimplify this process by overlooking therapeutic methods, directly mapping symptoms to herbs or syndromes to herbs, resulting in information loss and reducing the effectiveness of recommended prescriptions. Furthermore, the implicit, sparse, and many-to-many relationships between syndromes and therapeutic methods, coupled with the nonlinear interactions between therapeutic methods and herbs, further hinder the modeling of the complete SD process. To address these challenges, we propose a novel four-partite graph paradigm that explicitly models the four key components of SD and their interactions, preserving critical information at each step and aligning more closely with clinicians' decision-making logic. Building on this, we develop SDPR, an SD-based prescription recommendation model comprising four modules aligned with all SD steps. Then, we integrated them into a multi-task learning framework to fully capture the progressive prescription process. To handle the implicit and complex relationships among syndromes, therapeutic methods, and herbs, we introduce a syndrome-induced pre-training strategy and a therapeutic method-aware contrastive learning framework. Extensive experiments on public and real-world datasets validate SDPR's effectiveness in herb recommendation and prescription retrieval, confirming the strength of the four-partite graph paradigm. Our broader goal is to advance the intelligent development of TCM in healthcare.

Traffic flow prediction plays a pivotal role in intelligent transportation systems. While previous efforts have made significant advances in modeling spatio-temporal dependencies, the traffic data collected by sensors in real-world scenarios is inherently limited, and the scarcity of data impedes model optimization. Moreover, it is difficult for existing models to capture the dynamic properties caused by changes in traffic patterns because they often use shared parameterization to model the time correlations of all time periods. To address these challenges, we propose a Spatio-Temporal Contrastive Learning-based Adaptive Graph Augmentation (STCL-AGA) framework for traffic flow prediction, which captures and adapts to evolving traffic patterns by employing dynamic graph adjustments and flow data masking. Specifically, our approach leverages spatio-temporal correlations captured through convolution operations. It improves adaptability through adaptive graph augmentation, which dynamically adjusts the graph structure and incorporates nodes flow masking based on traffic conditions. This approach achieves accurate traffic predictions by effectively integrating temporal and spatial data transformations. Experiments on five benchmark datasets demonstrate that STCL-AGA consistently outperforms various state-of-the-art baselines.

Soft prompt tuning is a widely studied parameter-efficient fine-tuning method.However, it has a clear drawback: many soft tokens must be inserted into the input sequences to guarantee downstream performance.As a result, soft prompt tuning is less considered than Low-rank adaptation (LoRA) in the large language modeling (LLM) era.In this work, we propose a novel prompt tuning method, Instruction-Aware Prompt Tuning (IAPT), that requires only four soft tokens.First, we install a parameter-efficient soft prompt generator at each Transformer layer to generate idiosyncratic soft prompts for each input instruction.The generated soft prompts can be seen as a semantic summary of the input instructions and can effectively guide the output generation.Second, the soft prompt generators are modules with a bottleneck architecture consisting of a self-attention pooling operation, two linear projections, and an activation function.Pilot experiments show that prompt generators at different Transformer layers require different activation functions.Thus, we propose to learn the idiosyncratic activation functions for prompt generators automatically with the help of rational functions.We have conducted experiments on various tasks, and the experimental results demonstrate that (a) our IAPT method can outperform the recent baselines with comparable tunable parameters.(b) Our IAPT method is more efficient than LoRA under the singlebackbone multi-tenant setting.

Medical decision rules play a key role in many clinical decision support systems (CDSS).However, these rules are conventionally constructed by medical experts, which is expensive and hard to scale up.In this study, we explore the automatic extraction of medical decision rules from text, leading to a solution to construct large-scale medical decision rules.We adopt a formulation of medical decision rules as binary trees consisting of condition/decision nodes.Such trees are referred to as medical decision trees and we introduce several generative models to extract them from text.The proposed models inherit the merit of two categories of successful natural language generation frameworks, i.e., sequence-to-sequence generation and autoregressive generation.To unleash the potential of pretrained language models, we design three styles of linearization (natural language, augmented natural language and JSON code), acting as the target sequence for our models.Our final system achieves 67% tree accuracy on a comprehensive Chinese benchmark, outperforming state-of-the-art baseline by 12%.The result demonstrates the effectiveness of generative models on explicitly modeling structural decision-making roadmaps, and shows great potential to boost the development of CDSS and explainable AI.Our code will be open-source upon acceptance.

With the increasing number of web documents, the demand for translation has increased dramatically. Non-autoregressive translation (NAT) models can significantly reduce decoding latency to meet the growing translation needs, but they sacrifice translation quality. And there is still an irreparable performance gap between NAT models and strong autoregressive translation (AT) models at the corpus level. However, more fine-grained comparative experiments on AT and NAT are currently lacking. Therefore, in this paper, we first conducted analysis experiments at the sentence level and found complementarity and high similarity between the translations generated by AT and NAT. Then, based on this observation, we propose a general and effective method called NAT4AT, which can not only use NAT to speed up the inference speed of AT significantly but also improve its final translation quality. Specifically, NAT4AT first uses a NAT model to generate an original translation in parallel and then uses an AT model as a correction model to revise errors in the original translation. In this way, the AT model no longer needs to predict the entire translation but only needs to predict a small number of error parts in the NAT result. Extensive experimental results on major WMT benchmarks verify the generality and effectiveness of our method, whose translation quality is superior to the strong AT model and achieves a 5.0x speedup.

Xiaonan Li, Kai Lv, Hang Yan, Tianyang Lin, Wei Zhu, Yuan Ni, Guotong Xie, Xiaoling Wang, Xipeng Qiu. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2023.

Contrastive learning (CL) has achieved great success in various fields with self-supervised learning. However, CL under the supervised setting is not fully explored, especially how to utilize the class labels in CL. We propose a novel aligned contrastive finetuning (ACF) approach in this work. Specifically, we consider the label embeddings as labeled instances and put them in an InfoNCE loss objective together with the instance representations, thus aligning the label embeddings and instance representation in the same semantic space. In addition, we design a correlation-based regularization term to alleviate the anisotropy problem. Extensive experiments are conducted on language understanding and image classification tasks, demonstrating our ACF method’s competitiveness. ACF is off-the-shelf and can be plugged into any pre-trained models without additional network architectures or computation overhead.

Graph partitioning is crucial in distributed graph-parallel computing systems, and it is challenging for graph partitioning to optimize the communication cost and load balancing together. Existing state-of-the-art works, such as Powerlyra and TopoX, optimize the load balancing by randomly distributing the edges of high-degree vertices, which inevitably brings a high communication cost that is unbounded. This paper proposes a graph partition model that can minimize communication cost while maximizing load balancing. More specifically, we model the graph partition as the combinatorial design problem. Our proposed model can provide high-quality partition that guarantees that the computing load can be evenly distributed to each worker and minimizes the communication cost with a near-optimal theoretical boundary.Based on the proposed model, we extend the hybrid-cut partitioning algorithm for the power-law graph and propose HCPD, a hybrid-cut partitioning algorithm based on combinatorial design. HCPD uses the proposed model to optimize the load balancing and communication cost simultaneously for high-degree vertices, and assigns the high-degree vertices and their low-degree neighbors to the same workers by label propagation to reduce the overall communication cost. In this way, we partition the low-degree and high-degree vertices holistically and further improve the partition quality, unlike Powerlyra and TopoX, which deal with the two parts independently. Our experiments show that HCPD outperforms Powerlyra on PageRank task by up to 2× faster on real-world power-law graphs with billions of edges.

Finding frequent subgraph patterns in a big graph is an important problem with many applications such as classifying chemical compounds and building indexes to speed up graph queries. Since this problem is NP-hard, some recent parallel systems have been developed to accelerate the mining. However, they often have a huge memory cost, very long running time, suboptimal load balancing, and possibly inaccurate results. In this paper, we propose an efficient system called T-FSM for parallel mining of frequent subgraph patterns in a big graph. T-FSM adopts a novel task-based execution engine design to ensure high concurrency, bounded memory consumption, and effective load balancing. It also supports a new anti-monotonic frequentness measure called Fraction-Score, which is more accurate than the widely used MNI measure. Our experiments show that T-FSM is orders of magnitude faster than SOTA systems for frequent subgraph pattern mining. Our system code has been released at https://github.com/lyuheng/T-FSM.

Fairness has been taken as a critical metric in machine learning models, which is considered as an important component of trustworthy machine learning. In this paper, we focus on obtaining fairness for popular link prediction tasks, which are measured by dyadic fairness. A novel pre-processing methodology is proposed to establish dyadic fairness through data repairing based on optimal transport theory. With the well-established theoretical connection between the dyadic fairness for graph link prediction and a conditional distribution alignment problem, the dyadic repairing scheme can be equivalently transformed into a conditional distribution alignment problem. Furthermore, an optimal transport-based dyadic fairness algorithm called DyadicOT is obtained by efficiently solving the alignment problem, satisfying flexibility and unambiguity requirements. The proposed DyadicOT algorithm shows superior results in obtaining fairness compared to other fairness methods on two benchmark graph datasets.

Non-autoregressive translation (NAT) model achieves a much faster inference speed than the autoregressive translation (AT) model because it can simultaneously predict all tokens during inference.However, its translation quality suffers from degradation compared to AT.And existing NAT methods only focus on improving the NAT model's performance but do not fully utilize it.In this paper, we propose a simple but effective method called "Candidate Soups," which can obtain high-quality translations while maintaining the inference speed of NAT models.Unlike previous approaches that pick the individual result and discard the remainders, Candidate Soups (CDS) can fully use the valuable information in the different candidate translations through model uncertainty.Extensive experiments on two benchmarks (WMT'14 EN-DE and WMT'16 EN-RO) demonstrate the effectiveness and generality of our proposed method, which can significantly improve the translation quality of various base models.More notably, our best variant outperforms the AT model on three translation tasks with 7.6× speedup. 1

Session-based Recommendation (SR) aims to predict the next item for recommendation based on previously recorded sessions of user interaction. The majority of existing approaches to SR focus on modeling the transition patterns of items. In such models, the so-called micro-behaviors describing how the user locates an item and carries out various activities on it (e.g., click, add-to-cart, and read-comments), are simply ignored. A few recent studies have tried to incorporate the sequential patterns of micro-behaviors into SR models. However, those sequential models still cannot effectively capture all the inherent interdependencies between micro-behavior operations. In this work, we aim to investigate the effects of the micro-behavior information in SR systematically. Specifically, we identify two different patterns of micro-behaviors: “sequential patterns” and “dyadic relational patterns”. To build a unified model of user micro-behaviors, we first devise a multigraph to aggregate the sequential patterns from different items via a graph neural network, and then utilize an extended self-attention network to exploit the pair-wise relational patterns of micro-behaviors. Extensive experiments on three public real-world datasets show the superiority of the proposed approach over the state-of-the-art baselines and confirm the usefulness of these two different micro-behavior patterns for SR.

In online shopping, ever-changing fashion trends make merchants need to prepare more differentiated products to meet the diversified demands, and e-commerce platforms need to capture the market trend with a prophetic vision. For the trend prediction, the attribute tags, as the essential description of items, can genuinely reflect the decision basis of consumers. However, few existing works explore the attribute trend in the specific community for e-commerce. In this paper, we focus on the community trend prediction on the item attribute and propose a unified framework that combines the dynamic evolution of two graph patterns to predict the attribute trend in a specific community. Specifically, we first design a community-attribute bipartite graph at each time step to learn the collaboration of different communities. Next, we transform the bipartite graph into a hypergraph to exploit the associations of different attribute tags in one community. Lastly, we introduce a dynamic evolution component based on the recurrent neural networks to capture the fashion trend of attribute tags. Extensive experiments on three real-world datasets in a large e-commerce platform show the superiority of the proposed approach over several strong alternatives and demonstrate the ability to discover the community trend in advance.

Traditional Chinese Medicine (TCM) has the longest clinical history in Asia and contributes a lot to health maintenance worldwide. An essential step during the TCM diagnostic process is syndrome induction, which comprehensively analyzes the symptoms and generates an overall summary of the symptoms. Given a set of symptoms, the existing herb recommenders aim to generate the corresponding herbs as a treatment by inducing the implicit syndrome representations based on TCM prescriptions. As different symptoms have various importance during the comprehensive consideration, we argue that treating the co-occurred symptoms equally to do syndrome induction in the previous studies will lead to the coarse-grained syndrome representation. In this paper, we bring the attention mechanism to model the syndrome induction process. Given a set of symptoms, we leverage an attention network to discriminate the symptom importance and adaptively fuse the symptom embeddings. Besides, we introduce a TCM knowledge graph to enrich the input corpus and improve the quality of representation learning. Further, we build a KG-enhanced Multi-Graph Neural Network architecture, which performs the attentive propagation to combine node feature and graph structural information. Extensive experimental results on two TCM data sets show that our proposed model has the outstanding performance over the state-of-the-arts.

Given a data matrix \( D \) , a submatrix \( S \) of \( D \) is an order-preserving submatrix (OPSM) if there is a permutation of the columns of \( S \) , under which the entry values of each row in \( S \) are strictly increasing. OPSM mining is widely used in real-life applications such as identifying coexpressed genes and finding customers with similar preference. However, noise is ubiquitous in real data matrices due to variable experimental conditions and measurement errors, which makes conventional OPSM mining algorithms inapplicable. No previous work on OPSM has ever considered uncertain value intervals using the well-established possible world semantics. We establish two different definitions of significant OPSMs based on thepossible world semantics: (1) expected support-based and (2) probabilistic frequentness-based. An optimized dynamic programming approach is proposed to compute the probability that a row supports a particular column permutation, with a closed-form formula derived to efficiently handle the special case of uniform value distribution and an accurate cubic spline approximation approach that works well with any uncertain value distributions. To efficiently check the probabilistic frequentness, several effective pruning rules are designed to efficiently prune insignificant OPSMs; two approximation techniques based on the Poisson and Gaussian distributions, respectively, are proposed for further speedup. These techniques are integrated into our two OPSM mining algorithms, based on prefix-projection and Apriori, respectively. We further parallelize our prefix-projection-based mining algorithm using PrefixFPM, a recently proposed framework for parallel frequent pattern mining, and we achieve a good speedup with the number of CPU cores. Extensive experiments on real microarray data demonstrate that the OPSMs found by our algorithms have a much higher quality than those found by existing approaches.

Session-based Recommendations recommend the next possible item for the user with anonymous sessions, whose challenge is that the user’s behavioral preference can only be analyzed in a limited sequence to meet their need. Recent advances evaluate the effectiveness of the attention mechanism in the session-based recommendation. However, two simplifying assumptions are made by most of these attention-based models. One is to regard the last-click as the query vector to denote the user’s current preference, and the other is to consider that all items within the session are favorable for the final result, including the effect of unrelated items (i.e., spurious user behaviors). In this paper, we propose a novel Dual Sparse Attention Network for the session-based recommendation called DSAN to address these shortcomings. In this proposed method, we explore a learned target item embedding to model the user’s current preference and apply an adaptively sparse transformation function to eliminate the effect of the unrelated items. Experimental results on two real public datasets show that the proposed method is superior to the state-of-the-art session-based recommendation algorithm in all tests and also demonstrate that not all actions within the session are useful. To make our results reproducible, we have published our code on https://github.com/SamHaoYuan/DSANForAAAI2021.

In recent years, there are a large number of recommendation algorithms proposed in the literature, from traditional collaborative filtering to deep learning algorithms. However, the concerns about how to standardize open source implementation of recommendation algorithms continually increase in the research community. In the light of this challenge, we propose a unified, comprehensive and efficient recommender system library called RecBole (pronounced as [rEk'[email protected]]), which provides a unified framework to develop and reproduce recommendation algorithms for research purpose. In this library, we implement 73 recommendation models on 28 benchmark datasets, covering the categories of general recommendation, sequential recommendation, context-aware recommendation and knowledge-based recommendation. We implement the RecBole library based on PyTorch, which is one of the most popular deep learning frameworks. Our library is featured in many aspects, including general and extensible data structures, comprehensive benchmark models and datasets, efficient GPU-accelerated execution, and extensive and standard evaluation protocols. We provide a series of auxiliary functions, tools, and scripts to facilitate the use of this library, such as automatic parameter tuning and break-point resume. Such a framework is useful to standardize the implementation and evaluation of recommender systems. The project and documents are released at https://recbole.io/.

In this work, we propose a novel framework, Gradient Aligned Mutual Learning BERT (GAML-BERT), for improving the early exiting of BERT.GAML-BERT's contributions are two-fold.We conduct a set of pilot experiments, which shows that mutual knowledge distillation between a shallow exit and a deep exit leads to better performances for both.From this observation, we use mutual learning to improve BERT's early exiting performances, that is, we ask each exit of a multi-exit BERT to distill knowledge from each other.Second, we propose GA, a novel training method that aligns the gradients from knowledge distillation to cross-entropy losses.Extensive experiments are conducted on the GLUE benchmark, which shows that our GAML-BERT can significantly outperform the state-of-the-art (SOTA) BERT early exiting methods.

With the development of e-commerce platforms, e-commerce recommendation systems are playing an increasingly important role for the purpose of product recommendation. As a new attack model against e-commerce recommendation systems, the "Ride Item's Coattails" attack creates fake click information to establish the deceptive correlation between popular products and low-quality products in order to mislead the recommendation system of e-commerce platform to boost the sales of low-quality products. This attack is characterized by high concealment and strong destructiveness, which can cause great damage to e-commerce recommendation systems, and adversely affect the usability of the e-commerce platform and users' shopping experience. It is therefore of great practical significance to study how to quickly and effectively identify the false click information and the corresponding "Ride Item's Coattails" attack to better safeguard e-commerce recommendation systems. At present, there is no previously reported relevant research work conducted specifically for addressing the detection of the "Ride Item's Coattails" attack. In this work, we carried out pioneering work in analyzing and summarizing the characteristics of the false click information produced by attackers on the target products in the "Ride Item's Coattails" attack and designed a set of attack detection techniques suitable for e-commerce recommendation systems. Experimental results on real e-commerce datasets show that our proposed techniques can quickly and effectively detect the large-scale fake click information as well as the associated "Ride Item's Coattails" attack in e-commerce recommendation systems.

Frequent pattern mining (FPM) has been a focused theme in data mining research for decades, but there lacks a general programming framework that can be easily customized to mine different kinds of frequent patterns, and existing solutions to FPM over big transaction databases are IO-bound rendering CPU cores underutilized even though FPM is NP-hard. This paper presents, PrefixFPM, a general-purpose framework for FPM that is able to fully utilize the CPU cores in a multicore machine. PrefixFPM follows the idea of prefix projection to partition the workloads of PFM into independent tasks by divide and conquer. PrefixFPM exposes a unified programming interface to users who can customize it to mine their desired patterns, and the parallel execution engine is transparent to end-users and can be reused for mining all kinds of patterns. We have adapted the state-of-the-art serial algorithms for mining frequent patterns including subsequences, subtrees, and subgraphs on top of PrefixFPM, and extensive experiments demonstrate an excellent speedup ratio of PrefixFPM with the number of cores. A demo is available at https://youtu.be/PfioC0GDpsw; the code is available at https://github.com/yanlab19870714/PrefixFPM.

Herb recommendation plays a crucial role in the therapeutic process of Traditional Chinese Medicine (TCM), which aims to recommend a set of herbs to treat the symptoms of a patient. While several machine learning methods have been developed for herb recommendation, they are limited in modeling only the interactions between herbs and symptoms, and ignoring the intermediate process of syndrome induction. When performing TCM diagnostics, an experienced doctor typically induces syndromes from the patient's symptoms and then suggests herbs based on the induced syndromes. As such, we believe the induction of syndromes - an overall description of the symptoms - is important for herb recommendation and should be properly handled. However, due to the ambiguity and complexity of syndrome induction, most prescriptions lack the explicit ground truth of syndromes. In this paper, we propose a new method that takes the implicit syndrome induction process into account for herb recommendation. Specifically, given a set of symptoms to treat, we aim to generate an overall syndrome representation by effectively fusing the embeddings of all the symptoms in the set, so as to mimic how a doctor induces the syndromes. Towards symptom embedding learning, we additionally construct a symptom-symptom graph from the input prescriptions for capturing the relations (cooccurred patterns) between symptoms; we then build graph convolution networks (GCNs) on both symptom-symptom and symptom-herb graphs to learn symptom embedding. Similarly, we construct a herb-herb graph and build GCNs on both herbherb and symptom-herb graphs to learn herb embedding, which is finally interacted with the syndrome representation to predict the scores of herbs. The advantage of such a Multi-Graph GCN architecture is that more comprehensive representations can be obtained for symptoms and herbs. We conduct extensive experiments on a public TCM dataset, demonstrating significant improvements over state-of-the-art herb recommendation methods. Further studies justify the effectiveness of our design of syndrome representation and multiple graphs.

Session-based recommendation, which aims to predict the next action of an anonymous user base on the interaction information in a session, plays a crucial role in many online services. Recent works solve the problem with the latest deep learning techniques and have achieved good performance on some datasets. However, they have some shortcomings that affect their practical application value: a) the drift process of users' interests in the browsing is not well explored; b) the association between a user's current interests and general preferences in the session is not adequately considered. They mostly assume that the last interaction has a significant impact on the next interaction, which makes them work well only in limited scenarios and specific datasets. To address these limitations, we propose a session-based recommendation model called POEM, which explicitly considers the impact of interaction order relationships on recommendations by emphasizing position attributes in the session. Specifically, POEM models the macro and micro importance of each item in the session, the influence of user interaction order on the item-level collaboration, and the session-level collaboration reflected in the user interest drift process, respectively. Extensive experiments of the effectiveness, efficiency, and universality on three real-world datasets show that our method outperforms various state-of-the-art session-based recommendation methods consistently.

Personalized tag recommender systems suggest a list of tags to a user when he or she wants to annotate an item. They utilize users’ preferences and the features of items. Tensorfactorization techniques have been widely used in tag recommendation. Given the user-item pair, although the classic PITF (Pairwise Interaction Tensor Factorization) explicitly models the pairwise interactions among users, items and tags, it overlooks users’ short-term interests and suffers from data sparsity. On the other hand, given the user-item-time triple, time-aware approaches like BLL (Base-Level Learning) utilize the time effect to capture the temporal dynamics and the most popular tags on items to handle cold start situation of new users. However, it works only on individual level and the target resource level, which cannot find users’ potential interests. In this paper, we propose an unified tag recommendation approach by considering both time awareness and personalization aspects, which extends PITF by adding weightsto user-tag interaction and item-tag interaction respectively. Compared to PITF, our proposed model can depict temporal factor by temporal weights and relieve data sparsity problem by referencing the most popular tags on items. Further, our model brings collaborative filtering (CF) to time-aware models, which can mine information from global data and help improving the ability of recommending new tags. Different from the power-form functions used in the existing time aware recommendation models, we use the Hawkes process with the exponential intensity function to improve the model’s efficiency. The experimental results show that our proposed model outperforms the state of the art tag recommendation methods in accuracy and has better ability to recommend new tags.

Window functions have been a part of the SQL standard since 2003 and have been studied extensively during the past decade. They are widely used in data analysis; almost all the current mainstream commercial databases support window functions. However, in recent years the size of datasets is growing steeply; the existing window function implementations are not efficient enough. Recently, some sampling-based algorithms (e.g., online aggregation) are proposed to deal with large and complex data in relational databases, which offer us a flexible trade-off between accuracy and efficiency. However, few sampling techniques has been considered for window functions in databases. In this paper, we extend our previous work (Song et al. in Asia-Pacific web and web-age information management joint conference on web and big data, Springer, pp 229–244, 2017 ) and proposed two new algorithms: range-based global sampling algorithm and row-labeled sampling algorithm. The proposed algorithms use global sampling rather than local sampling and are more efficient than other existing algorithms. And we find our proposed algorithms out performed the baseline method over the TPC-H benchmark dataset.

In data-to-text Natural Language Generation (NLG) systems, computers need to find the right words to describe phenomena seen in the data. This paper focuses on the problem of choosing appropriate verbs to express the direction and magnitude of a percentage change (e.g., in stock prices). Rather than simply using the same verbs again and again, we present a principled data-driven approach to this problem based on Shannon’s noisy-channel model so as to bring variation and naturalness into the generated text. Our experiments on three large-scale real-world news corpora demonstrate that the proposed probabilistic model can be learned to accurately imitate human authors’ pattern of usage around verbs, outperforming the state-of-the-art method significantly.

It is an important problem in computational advertising to study the effects of different advertising channels upon user conversions, as advertisers can use the discoveries to plan or optimize advertising campaigns. In this paper, we propose a novel Probabilistic Multi-Touch Attribution (PMTA) model which takes into account not only which ads have been viewed or clicked by the user but also when each such interaction occurred. Borrowing the techniques from survival analysis, we use the Weibull distribution to describe the observed conversion delay and use the hazard rate of conversion to measure the influence of an ad exposure. It has been shown by extensive experiments on a large real-world dataset that our proposed model is superior to state-of-the-art methods in both conversion prediction and attribution analysis. Furthermore, a surprising research finding obtained from this dataset is that search ads are often not the root cause of final conversions but just the consequence of previously viewed ads.

Outsourcing keys (including passwords and data encryption keys) to professional password managers (honest-butcurious service providers) is attracting more and more attention from the researchers and users in the era of cloud computing. However, existing solutions in traditional data outsourcing scenario are unable to simultaneously meet the following three security requirements for keys outsourcing: (1) Confidentiality and privacy of keys; (2) Search privacy on identity attributes tied to keys; (3) Owner controllable authorization over his/her shared keys. In this paper, we propose CloudKeyBank, the first unified key management framework that addresses all the three goals above. To implement CloudKeyBank efficiently, we propose a new cryptographic primitive named Searchable Conditional Proxy Re-Encryption (SC-PRE) which combines the techniques of Hidden Vector Encryption (HVE) and Proxy Re-Encryption (PRE) seamlessly.

How can we know whether one classifier is really better than the other? In the area of text classification, since the publication of Yang and Liu's seminal SIGIR-1999 paper, it has become a standard practice for researchers to apply null-hypothesis significance testing (NHST) on their experimental results in order to establish the superiority of a classifier. However, such a frequentist approach has a number of inherent deficiencies and limitations, e.g., the inability to accept the null hypothesis (that the two classifiers perform equally well), the difficulty to compare commonly-used multivariate performance measures like F1 scores instead of accuracy, and so on. In this paper, we propose a novel Bayesian approach to the performance comparison of text classifiers, and argue its advantages over the traditional frequentist approach based on t-test etc. In contrast to the existing probabilistic model for F1 scores which is unpaired, our proposed model takes the correlation between classifiers into account and thus achieves greater statistical power. Using several typical text classification algorithms and a benchmark dataset, we demonstrate that the our approach provides rich information about the difference between two classifiers' performances.

Item recommendation helps people to discover their potentially interested items among large numbers of items. One most common application is to recommend top-n items on implicit feedback datasets (e.g., listening history, watching history or visiting history). In this paper, we assume that the implicit feedback matrix has local property, where the original matrix is not globally low rank but some sub-matrices are low rank. In this paper, we propose Local Weighted Matrix Factorization (LWMF) for top-n recommendation by employing the kernel function to intensify local property and the weight function to model user preferences. The problem of sparsity can also be relieved by sub-matrix factorization in LWMF, since the density of sub-matrices is much higher than the original matrix. We propose a heuristic method to select sub-matrices which approximate the original matrix well. The greedy algorithm has approximation guarantee of factor $$1-\frac{1}{e}$$ to get a near-optimal solution. The experimental results on two real datasets show that the recommendation precision and recall of LWMF are both improved about 30% comparing with the best case of weighted matrix factorization (WMF).

In modern cities, more and more people drive the vehicles, equipped with the GPS devices, which create a large scale of trajectories. Gathering and analyzing these large-scale trajectory data provide a new opportunity to understand the city dynamics and to reveal the hidden social and economic phenomena. This paper designs and implements a tool, named as PGWinFunc, to analyze trajectory data by extending a traditional relational database. Firstly we introduce some efficient query process and optimization methods for SQL Window Aggregate Functions in PostgreSQL. Secondly, we present how to mine the LBS (Location-Based Service) patterns, such as the average speed and traffic flow, from the large-scale trajectories with SQL expression with Window Aggregate Functions. Finally, the effectiveness and efficiency of the PGWinFunc tool are demonstrated and we also visualized the results by BAIDU MAP.

In multi-class text classification, the performance (effectiveness) of a classifier is usually measured by micro-averaged and macro-averaged F1 scores. However, the scores themselves do not tell us how reliable they are in terms of forecasting the classifier's future performance on unseen data. In this paper, we propose a novel approach to explicitly modelling the uncertainty of average F1 scores through Bayesian reasoning, and demonstrate that it can provide much more comprehensive performance comparison between text classifiers than the traditional frequentist null hypothesis significance testing (NHST).

Abstract Database as a service (DaaS), a new paradigm of software as a service based on cloud computing, is attracting more and more enterprises (data owners) to delegate their database management to a professional third party (database service provider) such as Amazon Web Services and Rackspace. Data owners in DaaS lose control of their sensitive data, which are stored in the delegated database and managed by the untrusted database service provider. Therefore, many encryption‐based approaches including attribute‐based encryption were proposed to implement fine‐grained access control in DaaS scenarios. However, most of the proposed access control enforcement approaches only support one or two of the following privacy guarantees: data privacy, policy privacy and key privacy. In this paper, we first propose a novel concept of DualAcE: a flexible fine‐grained dual access control enforcement mechanism in DaaS by efficiently combining the ciphertext‐policy attribute‐set‐based encryption with database service provider re‐encryption into a DaaS paradigm. The proposed mechanism has implemented dual access control enforcement with multi‐privacy guarantee: data privacy in delegated database, policy privacy in delegated authorization table and key privacy in key distribution process.We describe the security and efficiency analysis through cryptography theory and experimental results. Copyright © 2014 John Wiley & Sons, Ltd.

Bayesian network (BN) has been adopted as the underlying model for representing and inferring uncertain knowledge. As the basis of realistic applications centered on probabilistic inferences, learning a BN from data is a critical subject of machine learning, artificial intelligence, and big data paradigms. Currently, it is necessary to extend the classical methods for learning BNs with respect to data-intensive computing or in cloud environments. In this paper, we propose a parallel and incremental approach for data-intensive learning of BNs from massive, distributed, and dynamically changing data by extending the classical scoring and search algorithm and using MapReduce. First, we adopt the minimum description length as the scoring metric and give the two-pass MapReduce-based algorithms for computing the required marginal probabilities and scoring the candidate graphical model from sample data. Then, we give the corresponding strategy for extending the classical hill-climbing algorithm to obtain the optimal structure, as well as that for storing a BN by pairs. Further, in view of the dynamic characteristics of the changing data, we give the concept of influence degree to measure the coincidence of the current BN with new data, and then propose the corresponding two-pass MapReduce-based algorithms for BNs incremental learning. Experimental results show the efficiency, scalability, and effectiveness of our methods.

Numerous applications such as data integration, protein detection, and article copy detection share a similar core problem: given a string as the query, how to efficiently find all the similar answers from a large scale string collection. Many existing methods adopt a prefix-filter-based framework to solve this problem, and a number of recent works aim to use advanced filters to improve the overall search performance. In this paper, we propose a gram-based framework to achieve near maximum filter performance. The main idea is to judiciously choose the high-quality grams as the prefix of query according to their estimated ability to filter candidates. As this selection process is proved to be NP-hard problem, we give a cost model to measure the filter ability of grams and develop efficient heuristic algorithms to find high-quality grams. Extensive experiments on real datasets demonstrate the superiority of the proposed framework in comparison with the state-of-art approaches.

Explosive growth in the number of passwords for Web based applications and encryption keys for outsourced data storage well exceeds the management limit of users. Therefore, outsourcing keys (including passwords and data encryption keys) to professional password managers (honest-but-curious service providers) is attracting the attention of many users. However, existing solutions in a traditional data outsourcing scenario are unable to simultaneously meet the following three security requirements for keys outsourcing: (1) Confidentiality and privacy of keys; (2) Search privacy on identity attributes tied to keys; (3) Owner controllable authorization over his/her shared keys. In this paper, we propose CloudKeyBank, the first unified key management framework that addresses all the three goals above. Under our framework, the key owner can perform privacy and controllable authorization enforced encryption with minimum information leakage. To implement CloudKeyBank efficiently, we propose a new cryptographic primitive named Searchable Conditional Proxy Re-Encryption (SC-PRE) which combines the techniques of Hidden Vector Encryption (HVE) and Proxy Re-Encryption (PRE) seamlessly, and propose a concrete SC-PRE scheme based on existing HVE and PRE schemes. Our experimental results and security analysis show the efficiency and security goals are well achieved.

Detecting review spam is important for current e-commerce applications. However, the posted order of review has been neglected by the former work. In this paper, we explore the issue on fake review detection in review sequence, which is crucial for implementing online anti-opinion spam. We analyze the characteristics of fake reviews firstly. Based on review contents and reviewer behaviors, six time sensitive features are proposed to highlight the fake reviews. And then, we devise supervised solutions and a threshold-based solution to spot the fake reviews as early as possible. The experimental results show that our methods can identify the fake reviews orderly with high precision and recall.

With the convenient connection to network, more and more individual information including sensitive information, such as contact list in Mobile Phone or PDA, can be delegated to the professional third service provider to manage and maintain. The benefit of this paradigm is, on one hand to avoid the sensitive information leakage when individual devices failed or lost, on the other hand to make only the authorized users access and share the delegated information online anytime and anywhere. However, in this paradigm the critical problems to be resolved are to guarantee both the privacy of delegated individual information and the privacy of authorized users, and what is more important to afford the owners of communication devices to have high level of control and power to create their own particular access control policies. In this paper, we present an approach to implement the personalized access control at third service provider in a privacy preserving way. Our approach implements the critical problems above in this paradigm by using selective encryption, blind signature and the combination of role based access control and discretionary access control.

Database as a Service (DaaS) is a practical and useful paradigm, in which the Database Service Provider (DSP) hosts the delegated database generated from the Source DB of Data Owner (DO). Due to the untrusted DSP, most of the proposed approaches were concentrated on using encryption to guarantee the privacy of delegated database and using partition based index to speed up the query. However, few papers were proposed to guarantee the privacy of delegated access control policies. Therefore in order to improve the usability of delegated database and guarantee the privacy of delegated access control policies, a critical problem to be addressed in DaaS is to make the DSP enforce the delegated selective authorization policies correctly, but know nothing about the privacy of users or the privacy of delegated authorization policies. In this paper, we present a privacy preserving selective authorization enforcement approach to resolve the critical problem above. By using selective encryption, Pedersen commitment and access control policy polynomial, the privacy of delegated access control policies and the privacy of users can be efficiently guaranteed. Finally we analyze the security properties of our approach from different aspects.

Tracking data provenance (or lineage) has become increasingly important in many large-scale applications, and a few methods have been proposed to record data provenance recently. However, most of previous works mainly focus on deterministic databases except Trio style lineage that aims at probabilistic databases, which is much more challenging because of the exponential growth of possible world instances and dependence among intermediate tuples. This paper proposes an approach, named PHP-tree, to model how-provenance upon probabilistic databases. we also show how to evaluate probability based on a PHP-tree. Compared with Trio style lineage, our approach is independent of intermediate results and can calculate the probability both cases of restricted and complete propagation of data provenance. Detailed experimental results show the effectiveness, efficiency and scalability of our proposed model.

Cloud provides dynamically computing services for large scales of data over the Internet. IaaS(information as a service) is one of utilities to provide information service in Cloud computing. Large scales of XML data are produced continually in Internet. Efficient information filtering services are needed. Previous XML filter approaches aim at XPath queries. However, many users tend to use keywords to describe requirements. SLCA (Smallest Lowest Common Ancestor)-based XML keyword search is one of the most important information retrieval approaches. Former approaches focus on building centralized index for a large scale of XML document collection and can't process continuous XML streams. This paper addresses SLCA computing service for continuous XML document. A novel SLCA computing service is designed, where SLCAs are obtained in one scan of XML stream. We demonstrate the efficiency of our algorithms analytically and experimentally.

With the popular use of service-oriented technologies, database as a service (DaaS) as a new paradigm is becoming a more practical and useful model for those enterprises who can't afford the expensive DBMS products. However access control management by the database service provider (DSP) in the DaaS context is challenging because the DSP may be untrusted for the delegated data contents. In this paper, we first present an approach to implement the flexible access control enforcement management by applying a DSP re-encryption mechanism. Our approach not only can implement the selective access control of the encrypted data by the DSP, but also can relieve the users from the complex key derivation procedure. The underlying idea of our approach is that the DSP uses different re-encryption keys for users of the system to implement flexible access control enforcement management under the DSP re-encryption mechanism. We demonstrate the usefulness and security property of our flexible access control enforcement management, finally we analyze and resolve the possible attacks and information disclosure.

Web service has emerged as a new solution for enterprise application integration. Automatic service composite algorithm is to build a composite service for user's request or complex business requirement. Most former work uses the dependence among services, however the relationship is often complicated and the composition algorithm is time consuming. This paper proposes a type based service composition algorithm, which can calculate a feasible solution in linear time. We conduct a series of experiments with real dataset and generated data. The experimental results demonstrate the efficiency of proposed method.

With the increasing service-oriented applications, efficient service data management is becoming a practical and important issue. Currently, most exchange messages among different Web services is SOAP, which is represented by XML-formatted data. So, the main task of service data management is to manage XML format message files in order to provide efficient store and query for service requestors. Storing XML documents in relational databases has drawn much attention in recent years because it can leverage existing investments in relational database technologies. However, most former work defines relational schema for XML files based on heuristics without considering application characteristics, hence fails to produce efficient relational schema for various applications. In this paper, we propose an autonomic, workload-aware approach to generate relational schema according to dynamic application characteristics or user specified workload. Our approach adopts the genetic algorithm to find optimal mappings, and an elegant encoding method and related operations are proposed to manipulate mappings using bit strings. We implemented the proposed algorithm and our preliminary experiment results showed that our algorithm was more robust, more suitable for dynamic SOA applications and produced better relational schema than existing work

No abstract available.

Given a document repository, search engine is very helpful to retrieve information. Currently, vertical search is a hot topic, and Google Scholar [4] is an example for academic search. However, most vertical search engines only return the flat ranked list without an efficient result exhibition for given users. We study this problem and designed a vertical search engine prototype Dolphin, where the flexible user-oriented templates can be defined and the survey-like results are presented according to the template.

Recently, Web services have become a new technology trend for Enterprise Application Integration (EAI) and more and more applications based on Web services are emerging. One of the problems in using Web services in business applications such as logistics is services composition automatically and efficiently. In this paper, we present a Dynamic, Demand-Driven Web services Engine called D3D-Serv to implement composite service functionality that is used to dynamically build composite services from existing services according to different business logics and requirements. In this D3D-Serv framework, the most challenging function to implement is dynamic selection of service providers at run time. The highly dynamic and distributed nature of Web services often makes some service providers overloaded at certain times while others idle. To solve this problem, we propose an efficient services selection and execution strategy that is based on the queuing theory and can provide guarantees for the QOS (Quality of Service) under provider's limited resources. Preliminary experimental results have shown that this algorithm is effective.