Legal consultation question answering (Legal CQA) presents unique challenges compared to traditional legal QA tasks, including the scarcity of high-quality training data, complex task composition, and strong contextual dependencies.To address these, we construct JURISCQAD, a large-scale dataset of over 43,000 real-world Chinese legal queries annotated with expert-validated positive and negative responses, and design a structured task decomposition that converts each query into a legal element graph integrating entities, events, intents, and legal issues.We further propose JU-RISMA, a modular multi-agent framework supporting dynamic routing, statutory grounding, and stylistic optimization.Combined with the element graph, the framework enables strong context-aware reasoning, effectively capturing dependencies across legal facts, norms, and procedural logic.Trained on JURISCQAD and evaluated on a refined LawBench, our system significantly outperforms both general-purpose and legal-domain LLMs across multiple lexical and semantic metrics, demonstrating the benefits of interpretable decomposition and modular collaboration in Legal CQA.

The proliferation of misinformation on video-sharing platforms demands robust detection of video fake news. Existing methods struggle to integrate external world knowledge with internal multimodal cues, limiting their generalization and robustness. In this work, we propose TrueLens, a new framework for video fake news detection that gathers and consolidates dual-level evidence \zznotethrough three primary components, \ie, External Precedent Retriever, Adversarial Contrastor, and Internal Evidential Logic Fusion. At the external level, the External Precedent Retriever first decomposes the query video into textual, visual, and audio queries while leveraging multimodal large language models (MLLMs) to enhance the overall semantic representation. It then applies an entropy-guided multimodal retrieval mechanism to identify the two most similar reference videos from a gallery of real and fake samples, \ie, one real and one fake video. The Adversarial Contrastor integrates these references with the input video through contrastive attention, enhancing contextual reasoning. At the internal level, our Evidential Logic Fusion module aggregates multimodal signals from the Adversarial Contrastor to produce consistent, robust predictions. Extensive experiments on three benchmarks show that the proposed TrueLens consistently surpasses competitive baselines under both temporal and event settings by a clear margin, yielding up to a +21.60% F1 improvement under the event setting and achieving 93.73%, 90.64%, and 98.83% accuracy on the FakeSV, FakeTT, and FVC datasets under the temporal setting. The code for our project is available at https://github.com/JunyiChen-ai/TrueLens.

Academic question answering requires reasoning over heterogeneous scholarly graphs, where queries range from simple attribute lookups to multi-hop inference across author--paper--venue structures. Existing retrieval-augmented generation (RAG) systems struggle in this setting due to three limitations: (1) fixed retrieval strategies that do not adapt to varying query complexity, (2) the absence of sufficiency evaluation leading to incomplete or misaligned evidence, and (3) a lack of structured verification against graph facts. To address these issues, we propose an agentic heterogeneous graph RAG method that transforms the three core stages of the RAG pipeline into explicit agentic decision steps. A query-aware retrieval agent analyzes query type and selects an appropriate graph traversal strategy; a sufficiency-aware reranking agent assesses evidence completeness and adaptively expands the retrieved subgraph; and a graph-grounded verification agent checks entity, relation, and attribute correctness before finalizing the answer. Experiments on heterogeneous graphs constructed from OpenAlex and DBLP suggest that our method consistently outperforms strong LLM, graph-augmented RAG, and agent-based baselines.

As a key to accessing research impact, citation dynamics underpins research evaluation, scholarly recommendation, and the study of knowledge diffusion. Citation prediction is particularly critical for newborn papers, where early assessment must be performed without citation signals and under highly long-tailed distributions. We identify two key research gaps: (i) insufficient modeling of implicit factors of scientific impact, leading to reliance on coarse proxies; and (ii) a lack of bias-aware learning that can deliver stable predictions on lowly cited papers. We address these gaps by proposing a Bias-Aware Citation Prediction Framework, which combines multi-agent feature extraction with robust graph representation learning. First, a multi-agent x graph co-learning module derives fine-grained, interpretable signals, such as reproducibility, collaboration network, and text quality, from metadata and external resources, and fuses them with heterogeneous-network embeddings to provide rich supervision even in the absence of early citation signals. Second, we incorporate a set of robust mechanisms: a two-stage forward process that routes explicit factors through an intermediate exposure estimate, GroupDRO to optimize worst-case group risk across environments, and a regularization head that performs what-if analyses on controllable factors under monotonicity and smoothness constraints. Comprehensive experiments on two real-world datasets demonstrate the effectiveness of our proposed model. Specifically, our model achieves around a 13% reduction in error metrics (MALE and RMSLE) and a notable 5.5% improvement in the ranking metric (NDCG) over the baseline methods.

A well-designed layout is a prerequisite to any successful modular building. Beyond just room arrangement, a modular building layout (MBL) requires bundling multiple rooms with prefabricated walls into transportable modules, a process that is highly risky and costly without careful consideration from the outset. Traditional methods of developing MBLs rely heavily on rules-of-thumb inherited from cast-in-situ construction, but such practices overlooked emerging data-driven smart methods. This study presents HyperMB, a hypergraph representation and learning framework tailored for supporting MBL design decisions. It first collects layout drawings from constructed modular buildings and represents them as typed hypergraphs. A reproducible floorplan-to-hypergraph pipeline is then established by treating rooms and walls as first-class nodes and encoding “modularity” as higher-order hyperedges. The hypergraph learning model then aggregates both node- and hyperedge-level features. Empirical results show that it outperforms two graph baselines on the same tasks in both within-dataset and cross-dataset settings, especially at capturing complex inter- and intra-module dependencies. HyperMB can discover latent design patterns and high-risk logistical archetypes. In generative experiments, it synthesizes modularity hyperedges with 91% validity in simple layouts, while also revealing limitations in complex module configurations. By coupling with hypergraph learning, HyperMB provides practical decision support for MBL design and planning. It lays a foundation for future automation in modular building design.

Despite growing interest in graph-based models for knowledge recombination prediction using academic knowledge graphs, existing approaches suffer from significant limitations: they fail to learn informative and robust knowledge entity representations by neglecting high-order information, inadequately account for real-world dynamics, and crucially, cannot provide readable rationales for their predictions. We address these challenges with H2GLM, which reformulates traditional graph learning as heterogeneous hypergraph learning to capture high-order information, incorporating a variational autoencoder (VAE) mechanism to enhance informativeness and robustness. Our approach then integrates large language models (LLMs) with the learned graph contextual information through a step-wise methodology, enabling evidence-supported decisions with clear, readable rationales. Experimental results highlight that H2GLM outperforms previous strong graph-based and LLM-based baselines by 4% to 8% in accuracy, 3% to 9% in AUC and 5% to 8% in F1 on extensive academic knowledge graphs containing over 1,000,000 nodes, with a small amount of training data. Visualizations and case studies further illustrate our method’s substantial utility over existing approaches in real-world scenarios. Further explainability and efficiency analyses underscore the practical value of our method

OBJECTIVE: Precision medicine applications depend on elucidating the underlying molecular mechanisms of disease. However, many disorders, such as endometriosis, remain poorly characterised genetically due to data scarcity, positive-unlabelled (PU) imbalance, and the heterogeneous structure of biomedical knowledge. This study aims to develop HetBio-CLiP, a novel heterogeneous graph-based contrastive learning (CL) methodology that improves candidate gene prioritisation across heterogeneous biomedical graphs by explicitly addressing positive-unlabelled learning constraints and structural heterogeneity. METHODS: HetBio-CLiP (Heterogeneous Biomedical graph Contrastive Learning with Interactions and Positive-Unlabelled learning) integrates multi-relational genomic, variant, and clinical data from real-world patient cohorts. To address the challenges of data scarcity and class imbalance, the graph neural network (GNN) methodology combines heterogeneous graph CL with PU learning. Furthermore, the model incorporates an interpretable GNNShap explainer to provide transparency at both the feature and edge levels, to assess the biological relevance of the predictions. RESULTS: HetBio-CLiP achieved superior performance across key evaluation metrics, attaining the highest Area Under the Curve (AUC) of 0.9489 ± 0.04 and Area Under the Precision-Recall Curve (AUPR) of 0.9401 ± 0.05. This outperformed state-of-the-art baselines such as M-SAGEGraph (AUC 0.9149 ± 0.04) and GAT (AUC 0.9138 ± 0.07). In terms of ranking candidate genes, the model achieved the highest Precision@42 (0.6933 ± 0.06) and TP@42 (29.1 ± 2.4) and second-highest Normalised Discounted Cumulative Gain (NDCG@42) (0.827 ± 0.06). Ablation studies confirmed that the integration of PU and contrastive learning was essential for these performance gains. Biologically, the model successfully prioritised known pathophysiology drivers, with top-ranked validated gene predictions including WNT4, GREB1, and ESR1. CONCLUSION: HetBio-CLiP presents an effective and interpretable methodology for uncovering candidate gene-disease relationships in data-scarce environments. By combining CL with PU-training, the proposed method addresses critical limitations of current gene prioritisation models and produces biologically meaningful predictions. The model demonstrates a strong balance between ranking accuracy and recall, identifying broader sets of plausible candidates than existing methods. This approach benefits researchers, clinicians, and downstream genomic studies by offering an improved foundation for identifying novel disease-associated genes and guiding future experimental investigation in endometriosis and other complex diseases.

Deep neural networks have shown significant advancements in modelling complex non-linear relationships in high-dimensional biomedical data. Understanding the interplay between genetic variants and disease susceptibility is still a considerable challenge that prevents certain genomic diseases to be predicted accurately for clinical interventions. In this study, we introduce the Extensive Multi-Variant Deep Neural Network (EMV-DNN), an innovative deep learning methodology designed to enhance polygenic risk prediction. Unlike conventional polygenic risk score methods, EMV-DNN incorporates single nucleotide polymorphisms (SNPs) alongside structural variants including insertions and deletions (indels), short tandem repeats (STRs), and copy number variants (CNVs) using variant-specific subnetworks to extract informative embeddings which capture a richer and holistic genomic context. Evaluated on real-world cohorts from the UK Biobank and All of Us, EMV-DNN outperforms conventional PRS methods and classic machine learning algorithms across binary and multi-class prediction tasks. Beyond predictive performance, SHapley Additive exPlanations (SHAP) analysis revealed biologically plausible variant–gene–disease associations, highlighting pathways related to endometrial cell proliferation, fibrosis, and immune regulation. Our findings underscore the value of multi-variant integration and non-linear approaches to capture the intricate genetic architecture of complex genomic diseases. Despite challenges such as dataset limitations and the complexity of diseases with multiple contributing factors, the EMV-DNN methodology presents a promising avenue for enhancing the predictive accuracy of PRS, thereby facilitating personalized healthcare interventions and advancing our understanding of genetic predispositions to disease.

Abstract The classification of research according to its aims has been a longstanding focus in the fields of quantitative science studies and R&D statistics. Since 1963, the Organization for Economic Co‐operation and Development (OECD) has employed a classical distinction among basic, applied, and experimental research. Building on this framework, our previous work highlighted the utility of differentiating between scientific and societal progress as two primary research objectives. This distinction enabled the quantitative analysis of scientific publication abstracts and the development of an automated method for large‐scale classification. In the current study, we systematically evaluate text classification techniques, including traditional text mining models, classification tools, BERT‐based language models, and decoder‐only large language models (LLMs) such as ChatGPT. Our findings show that the fine‐tuned GPT‐4o‐mini model performs the best among single‐model approaches. However, traditional and BERT‐based models outperform in certain fine‐grained classification tasks. Leveraging majority voting strategies to incorporate their strengths yields performance comparable to closed‐source GPT models. A case study on 10 biomedical journals further validates the method, demonstrating strong alignment between journal scopes, model predictions, and outputs generated by the fine‐tuned GPT‐4o‐mini model. These results highlight the robustness and practical effectiveness of the proposed methodology for nuanced research aim classification.

Cancer risk prediction is a cornerstone of personalized medicine that offers opportunities for early detection and preventive interventions. However, the current models are designed to predict cancer risk face several challenges. First, most rely on traditional statistical methods, which struggle to capture the complexity of genetic, family medical history, and lifestyle factors. Hence, the accuracy of these models is limited. Additionally, the models neglect to integrate multidimensional data sources, particularly genetic information like single nucleotide polymorphisms (SNPs), which could enhance prediction accuracy. Third, while the system might effectively predict risk, it cannot translate those predictions into actionable healthcare recommendations to reduce cancer risk. In this study, we address all three of these limitations. With a focus on six prevalent cancers—we extracted SNP data from the UK Biobank and designed a novel risk prediction model for cancer and personalized healthcare recommendations based upon the mixture of experts (MoE) paradigm and large language models (LLMs), respectively. Named MoE-HRS, experts based two router networks for separate processing by the Transformer and the convolutional neural network (CNN). Experiments on UK Biobank data show that our model outperforms state-of-the-art cancer risk prediction models. To bridge the gap between risk prediction and practical healthcare applications, we devised a healthcare recommender system powered by LLMs. This approach holds promise for enhancing early detection rates and promoting preventive healthcare management (relevant coding and data are available at https://github.com/bjtu-lucas-nlp/MoE-HRS ).

Abstract Predicting a researcher's knowledge trajectories beyond their current foci can leverage potential inter‐/cross‐/multi‐disciplinary interactions to achieve exploratory innovation. In this study, we present a method of diffusion‐based network analytics for knowledge trajectory recommendation. The method begins by constructing a heterogeneous bibliometric network consisting of a co‐topic layer and a co‐authorship layer. A novel link prediction approach with a diffusion strategy is then used to capture the interactions between social elements (e.g., collaboration) and knowledge elements (e.g., technological similarity) in the process of exploratory innovation. This diffusion strategy differentiates the interactions occurring among homogeneous and heterogeneous nodes in the heterogeneous bibliometric network and weights the strengths of these interactions. Two sets of experiments—one with a local dataset and the other with a global dataset—demonstrate that the proposed method is prior to 10 selected baselines in link prediction, recommender systems, and upstream graph representation learning. A case study recommending knowledge trajectories of information scientists with topical hierarchy and explainable mediators reveals the proposed method's reliability and potential practical uses in broad scenarios.

The International Classification of Diseases (ICD) is a standard for categorizing physical conditions, which has been widely used for analyzing clinical data and monitoring health issues. Manual ICD coding takes a long time and is vulnerable to errors, so people pay more and more attention to the application of deep neural networks in ICD automatic coding. However, there is still no comprehensive review of these studies and prospects for further research. This paper is not limited to the study of deep neural networks, but gives a formal definition of ICD coding problems, and then systematically reviews the existing literature on how to design deep neural networks to address the four major challenges of ICD coding tasks. This paper also summarizes the public data sets and future research directions, to provide a guidance for the research of ICD coding in medical field.

Citation data, along with other bibliographic datasets, have long been adopted by the knowledge and data discovery community as an important direction for presenting the validity and effectiveness of proposed algorithms and strategies. Many top computer scientists are also excellent researchers in the science of science. The purpose of this workshop is to bridge the two communities (i.e., the knowledge discovery community and the science of science community) together as the scholarly activities become salient web and social activities that start to generate a ripple effect on broader knowledge discovery communities. This workshop will showcase the current data-driven science of science research by highlighting several studies and constructing a community of researchers to explore questions critical to the future of data-driven science of science, especially a community of data-driven science of science in Data Science so as to facilitate collaboration and inspire innovation. Through discussion on emerging and critical topics in the science of science, this workshop aims to help generate effective solutions for addressing environmental, societal, and technological problems in the scientific community.

Abstract Power dynamics influence every aspect of scientific collaboration. Team power dynamics can be measured by team power level and team power hierarchy. Team power level is conceptualized as the average level of the possession of resources, expertise, or decision‐making authorities of a team. Team power hierarchy represents the vertical differences of the possessions of resources in a team. In Science of Science, few studies have looked at scientific collaboration from the perspective of team power dynamics. This research examines how team power dynamics affect team impact to fill the research gap. In this research, all coauthors of one publication are treated as one team. Team power level and team power hierarchy of one team are measured by the mean and Gini index of career age of coauthors in this team. Team impact is quantified by citations of a paper authored by this team. By analyzing over 7.7 million teams from Science (e.g., Computer Science, Physics), Social Sciences (e.g., Sociology, Library & Information Science), and Arts & Humanities (e.g., Art), we find that flat team structure is associated with higher team impact, especially when teams have high team power level. These findings have been repeated in all five disciplines except Art, and are consistent in various types of teams from Computer Science including teams from industry or academia, teams with different gender groups, teams with geographical contrast, and teams with distinct size.

Whereas traditional science maps emphasize citation statistics and static relationships, this paper presents a term‐based method to identify and visualize the evolutionary pathways of scientific topics in a series of time slices. First, we create a data preprocessing model for accurate term cleaning, consolidating, and clustering. Then we construct a simulated data streaming function and introduce a learning process to train a relationship identification function to adapt to changing environments in real time, where relationships of topic evolution, fusion, death, and novelty are identified. The main result of the method is a map of scientific evolutionary pathways. The visual routines provide a way to indicate the interactions among scientific subjects and a version in a series of time slices helps further illustrate such evolutionary pathways in detail. The detailed outline offers sufficient statistical information to delve into scientific topics and routines and then helps address meaningful insights with the assistance of expert knowledge. This empirical study focuses on scientific proposals granted by the United States National Science Foundation, and demonstrates the feasibility and reliability. Our method could be widely applied to a range of science, technology, and innovation policy research, and offer insight into the evolutionary pathways of scientific activities.