arXiv subjects

Tao Jia

Publications and source records attributed to Tao Jia.

At least 19 recordsLinked to original sources

arXiv2026

NGM-RAG: Neural Graph Matching based Retrieval-Augmented Generation

Retrieval-Augmented Generation (RAG) significantly enhances the ability of Large Language Models (LLMs) to provide accurate and contextually relevant answers by dynamically integrating external databases. However, traditional RAG methods are primarily constrained by their reliance on text-based retrieval strategies, which often struggle with complex questions requiring multi-hop reasoning. To address this limitation, we introduce Neural Graph Matching based Retrieval-Augmented Generation (NGM-RAG), a novel framework that leverages graph structures to effectively capture and utilize relational knowledge for improved retrieval and answer generation. NGM-RAG explicitly incorporates graph construction, graph matching, and answer generation into a unified process. Within this framework, we propose a neural graph matching approach that combines text-based matching with Graph Neural Networks (GNNs). By employing an adaptive weighting strategy, NGM-RAG efficiently integrates multiple matching methods to select the most relevant contextual node information for answer generation. Experimental results on multi-hop question answering and long-context summarization tasks demonstrate that our NGM-RAG model achieves superior performance compared to both traditional NaiveRAG methods and state-of-the-art graph-enhanced approaches such as GraphRAG and LightRAG.

cs.IR↗

arXiv2026

CMSL: Constructive Multi-Sequence Learning for Recommendation Systems

Sequence learning has emerged as the promising paradigm in recommendation systems, surpassing traditional Deep Learning Recommendation Models (DLRM) by capturing the temporal nuances of user behavior. However, current state-of-the-art architectures operate under a limiting analogy: they treat user history as a monolithic chronological sequence like a sentence in a Large Language Model (LLM). We observe a fundamental divergence between natural language and recommendation data: unlike the linear, logical flow of text, user history is inherently multi-faceted. A user's journey is a fragmented reflection of diverse interests, resulting in much weaker coherence between items than is found in LLM training data. This lack of structural unity leads to context pollution. In single-sequence modeling, unrelated behaviors compete for the same attention budget. This "noisy" signal dilutes the model's focus, effectively capping its ability to discern high-intent patterns from background activity. To address this, we propose Constructive Multi-Sequence Learning (CMSL), a paradigm shift from passive sequence ingestion to active "context engineering" that constructs multiple coherent sequences in latent space. CMSL leverages a learnable Sequence Construction Module to disentangle user history into "pure" thematic strands, followed by a linear attention mechanism to efficiently model these strands at scale. CMSL has been deployed across ranking and retrieval tasks and across four major surfaces at Meta.

cs.IR↗

arXiv2026

Position: AI for Science Should Treat Measurement-to-Dataset Pipelines as Inference Components

AI for Science (AI4Science) workflows often treat the released dataset as a fixed interface to the underlying system. However, in domains relying on \emph{indirect observation}, the learner observes a derivative representation produced by multi-stage measurement, reconstruction, and preprocessing pipelines. \textbf{We argue that these measurement-to-dataset pipelines are inference components: treating their outputs as ``given data'' freezes an observation model and obscures uncertainty over feasible pipeline choices.} We identify three failure modes arising from this ``frozen lens'': \textbf{(C1) hidden hypothesis space}, where the released dataset does not specify the pipeline configuration or its validity conditions; \textbf{(C2) uncertified transportability}, where a pipeline may be documented but its regime of validity is untested, so failures under distribution shift cannot be adjudicated; \textbf{(C3) ungoverned multiplicity}, where many defensible pipelines exist and dispersion is real but not propagated into uncertainty-aware evidence. We stress-test these claims with a large-scale neuroscience empirical audit, finding a survival rate of $\approx 0.0004\%$ under a cross-dataset stability criterion. We call on the AI4Science community to make pipelines \emph{computable} inference objects via domain-specific Computable Observation Frameworks. This shift enables quantifying pipeline adequacy and stability, converting implicit implementation choices into auditable, reproducible, and cumulative scientific evidence.

cs.LG↗

arXiv2026

PDSL: Propagation Dynamics Aware Framework for Source Localization

Source localization is a representative inverse inference task in information propagation, aiming to identify the source node or node set that triggers the propagation results based on the observed information. A primary challenge is quantifying the inherent uncertainty between observed outcomes and potential sources. Although deep generative models have partially mitigated this issue, most existing approaches primarily focus on uncertainty induced by network topology, attempting to learn a direct mapping from propagation outcomes to sources based on network structure, while overlooking the additional uncertainty stemming from the highly stochastic nature of the propagation process. To address this limitation, we propose a Propagation Dynamics aware framework for Source Localization (PDSL), a novel method that integrates a deep generative model with propagation dynamics to approximate the source distribution and explicitly mitigate uncertainty arising from diffusion stochasticity. Moreover, we employ Graph Neural Ordinary Differential Equations to model the continuous dynamics of diffusion processes without relying on a predefined diffusion mechanism. Additionally, a matching mechanism is designed to extract relevant data blocks that enhance source generation reliability. Comprehensive experiments on both synthetic and real-world diffusion datasets demonstrate the superior performance of the proposed framework across diverse application scenarios.

cs.SI↗

arXiv2026

Inductive Dual-Polarity Modeling via Static-Dynamic Disentanglement for Dynamic Signed Networks

Dynamic signed networks (DSNs) are common in online platforms, where time-stamped positive and negative relations evolve over time. A core task in DSNs is dynamic edge prediction, which forecasts future relations by jointly modeling edge existence and polarity (positive, negative, or non-existent). However, existing dynamic signed network embedding (DSNE) methods often entangle positive and negative signals within a shared temporal state and rely on node-specific temporal trajectories, which can obscure polarity-asymmetric dynamics and harm inductive generalization, especially under cold-start evaluation. We study an inductive setting where each test edge contains at least one endpoint node held out from training, while its interactions prior to the prediction time are available as historical evidence. The model must therefore infer representations for unseen nodes solely from such limited history. We propose IDP-DSN, an Inductive Dual-Polarity framework for Dynamic Signed Networks. IDP-DSN maintains sign-selective memories to model positive and negative temporal dynamics separately, performs history-only neighborhood inference for unseen nodes (instead of learned node-wise trajectories), and enforces polarity-wise static--dynamic disentanglement via an orthogonality regularizer. Experiments on BitcoinAlpha, BitcoinOTC, Wiki-RfA, and Epinions demonstrate consistent improvements over the strongest baselines, achieving relative Macro-F1 gains of 16.8/23.4%, 16.9/24%, 30.1/25.5%, and 18.7/28.9% in the transductive/inductive settings, respectively. These results highlight the effectiveness of IDP-DSN on DSNs, particularly under inductive cold-start evaluation for dynamic signed edge prediction.

cs.SI↗

arXiv2026

Not Just Large: Tall Teams Dominate East Asia's Scientific Production

Purpose: This study compares the hierarchical structure of scientific teams across countries and investigates factors associated with the observed cross-national differences. Design/methodology/approach: Drawing on 150,817 publications with author contribution statements, we focus on the 15 countries with the largest volume of scientific publications, examine cross-country variations in the proportion of tall teams, and analyze how this proportion correlates with other factors. Findings: Scientific output from East Asia is dominated by tall teams, which persist after controlling for team size, indicating that this pattern cannot be fully accounted for by the prevalence of larger teams in these countries. Cultural factors, measured by Power Distance, as well as the observed funding patterns of major basic science agencies, are associated with the dominance of tall teams in East Asia. Research limitations: This study is limited by its reliance on publications with author contribution statements, which may introduce selection bias; its focus on cultural and funding factors, while leaving other institutional contexts unexamined; and its use of a leadership concentration measure that does not capture other dimensions of hierarchy. Practical implications: Understanding cross-national differences in research team structures and their associated cultural and institutional factors can inform science policy and team management. Originality/value: This study provides a systematic cross-national comparison of team hierarchy and offers a mechanistic understanding of the dominance of tall teams in East Asia, highlighting associations with cultural and funding factors.

cs.DL↗

arXiv2026

Superconductivity and Electron Correlations in Kagome Metal LuOs3B2

We report a comprehensive investigation of the physical properties of LuOs3B2, characterized by an ideal Os-based kagome lattice. Resistivity and magnetization measurements confirm the emergence of type-II bulk superconductivity with a critical temperature Tc=4.63 K. The specific heat jump and the calculated electron-phonon coupling parameter support a moderately coupled superconducting state. Electron correlation effects are supported by the enhanced Wilson ratios. First-principles calculations reveal hallmark features of kagome band structure, including Dirac points, van Hove singularities, and quasi-flat bands, primarily derived from the Os d orbitals. The inclusion of spin-orbit coupling opens a gap at the Dirac points, significantly altering the electronic properties. Furthermore, the superconductivity and electronic properties of isomorphic compounds are discussed. This work provides a thorough exploration of the superconducting and normal states of LuOs3B2, deepening the understanding of kagome superconductors.

cond-mat.supr-con↗

arXiv2026

Accelerating Benchmarking of Functional Connectivity Modeling via Structure-aware Core-set Selection

Benchmarking the hundreds of functional connectivity (FC) modeling methods on large-scale fMRI datasets is critical for reproducible neuroscience. However, the combinatorial explosion of model-data pairings makes exhaustive evaluation computationally prohibitive, preventing such assessments from becoming a routine pre-analysis step. To break this bottleneck, we reframe the challenge of FC benchmarking by selecting a small, representative core-set whose sole purpose is to preserve the relative performance ranking of FC operators. We formalize this as a ranking-preserving subset selection problem and propose Structure-aware Contrastive Learning for Core-set Selection (SCLCS), a self-supervised framework to select these core-sets. SCLCS first uses an adaptive Transformer to learn each sample's unique FC structure. It then introduces a novel Structural Perturbation Score (SPS) to quantify the stability of these learned structures during training, identifying samples that represent foundational connectivity archetypes. Finally, while SCLCS identifies stable samples via a top-k ranking, we further introduce a density-balanced sampling strategy as a necessary correction to promote diversity, ensuring the final core-set is both structurally robust and distributionally representative. On the large-scale REST-meta-MDD dataset, SCLCS preserves the ground-truth model ranking with just 10% of the data, outperforming state-of-the-art (SOTA) core-set selection methods by up to 23.2% in ranking consistency (nDCG@k). To our knowledge, this is the first work to formalize core-set selection for FC operator benchmarking, thereby making large-scale operators comparisons a feasible and integral part of computational neuroscience. Code is publicly available on https://github.com/lzhan94swu/SCLCS

cs.LG↗

arXiv2026

LIR$^3$AG: A Lightweight Rerank Reasoning Strategy Framework for Retrieval-Augmented Generation

Retrieval-Augmented Generation (RAG) effectively enhances Large Language Models (LLMs) by incorporating retrieved external knowledge into the generation process. Reasoning models improve LLM performance in multi-hop QA tasks, which require integrating and reasoning over multiple pieces of evidence across different documents to answer a complex question. However, they often introduce substantial computational costs, including increased token consumption and inference latency. To better understand and mitigate this trade-off, we conduct a comprehensive study of reasoning strategies for reasoning models in RAG multi-hop QA tasks. Our findings reveal that reasoning models adopt structured strategies to integrate retrieved and internal knowledge, primarily following two modes: Context-Grounded Reasoning, which relies directly on retrieved content, and Knowledge-Reconciled Reasoning, which resolves conflicts or gaps using internal knowledge. To this end, we propose a novel Lightweight Rerank Reasoning Strategy Framework for RAG (LiR$^3$AG) to enable non-reasoning models to transfer reasoning strategies by restructuring retrieved evidence into coherent reasoning chains. LiR$^3$AG significantly reduce the average 98% output tokens overhead and 58.6% inferencing time while improving 8B non-reasoning model's F1 performance ranging from 6.2% to 22.5% to surpass the performance of 32B reasoning model in RAG, offering a practical and efficient path forward for RAG systems.

cs.CL↗

arXiv2025

A generalized motif-based Naïve Bayes model for sign prediction in complex networks

Signed networks, encoding both positive and negative interactions, are essential for modeling complex systems in social and financial domains. Sign prediction, which infers the sign of a target link, has wide-ranging practical applications. Traditional motif-based Naïve Bayes models assume that all neighboring nodes contribute equally to a target link's sign, overlooking the heterogeneous influence among neighbors and potentially limiting performance. To address this, we propose a generalizable sign prediction framework that explicitly models the heterogeneity. Specifically, we design two role functions to quantify the differentiated influence of neighboring nodes. We further extend this approach from a single motif to multiple motifs via two strategies. The generalized multiple motifs-based Naïve Bayes model linearly combines information from diverse motifs, while the Feature-driven Generalized Motif-based Naïve Bayes (FGMNB) model integrates high-dimensional motif features using machine learning. Extensive experiments on four real-world signed networks show that FGMNB consistently outperforms five state-of-the-art embedding-based baselines on three of these networks. Moreover, we observe that the most predictive motif structures differ across datasets, highlighting the importance of local structural patterns and offering valuable insights for motif-based feature engineering. Our framework provides an effective and theoretically grounded solution to sign prediction, with practical implications for enhancing trust and security in online platforms.

cs.CR↗

arXiv2025

Knowledge Independence Breeds Disruption but Limits Recognition

Despite extensive research on scientific disruption, two questions remain: why disruption has declined amid growing knowledge, and why disruptive work receives fewer and delayed citations. One way to address these questions is to identify an intrinsic, paper-level property that reliably predicts disruption and explains both patterns. Here, we propose a novel measure, knowledge independence, capturing the extent to which a paper draws on references that do not cite one another. Analyzing 114 million publications, we find that knowledge independence strongly predicts disruption and mediates the disruptive advantage of small, onsite, and fresh teams. Its long-term decline, nonreproducible by null models, provides a mechanistic explanation for the parallel decline in disruption. Causal and simulation evidence further indicates that knowledge independence drives the persistent trade-off between disruption and impact. Taken together, these findings fill a critical gap in understanding scientific innovation, revealing a universal law: Knowledge independence breeds disruption but limits recognition.

physics.soc-ph↗

arXiv2025

Learning to Compress Graphs via Dual Agents for Consistent Topological Robustness Evaluation

As graph-structured data grow increasingly large, evaluating their robustness under adversarial attacks becomes computationally expensive and difficult to scale. To address this challenge, we propose to compress graphs into compact representations that preserve both topological structure and robustness profile, enabling efficient and reliable evaluation. We propose Cutter, a dual-agent reinforcement learning framework composed of a Vital Detection Agent (VDA) and a Redundancy Detection Agent (RDA), which collaboratively identify structurally vital and redundant nodes for guided compression. Cutter incorporates three key strategies to enhance learning efficiency and compression quality: trajectory-level reward shaping to transform sparse trajectory returns into dense, policy-equivalent learning signals; prototype-based shaping to guide decisions using behavioral patterns from both high- and low-return trajectories; and cross-agent imitation to enable safer and more transferable exploration. Experiments on multiple real-world graphs demonstrate that Cutter generates compressed graphs that retain essential static topological properties and exhibit robustness degradation trends highly consistent with the original graphs under various attack scenarios, thereby significantly improving evaluation efficiency without compromising assessment fidelity.

cs.LG↗

arXiv2025

NCSAC: Effective Neural Community Search via Attribute-augmented Conductance

Identifying locally dense communities closely connected to the user-initiated query node is crucial for a wide range of applications. Existing approaches either solely depend on rule-based constraints or exclusively utilize deep learning technologies to identify target communities. Therefore, an important question is proposed: can deep learning be integrated with rule-based constraints to elevate the quality of community search? In this paper, we affirmatively address this question by introducing a novel approach called Neural Community Search via Attribute-augmented Conductance, abbreviated as NCSAC. Specifically, NCSAC first proposes a novel concept of attribute-augmented conductance, which harmoniously blends the (internal and external) structural proximity and the attribute similarity. Then, NCSAC extracts a coarse candidate community of satisfactory quality using the proposed attribute-augmented conductance. Subsequently, NCSAC frames the community search as a graph optimization task, refining the candidate community through sophisticated reinforcement learning techniques, thereby producing high-quality results. Extensive experiments on six real-world graphs and ten competitors demonstrate the superiority of our solutions in terms of accuracy, efficiency, and scalability. Notably, the proposed solution outperforms state-of-the-art methods, achieving an impressive F1-score improvement ranging from 5.3\% to 42.4\%. For reproducibility purposes, the source code is available at https://github.com/longlonglin/ncsac.

cs.SI↗

arXiv2025

MIARec: Mutual-influence-aware Heterogeneous Network Embedding for Scientific Paper Recommendation

With the rapid expansion of scientific literature, scholars increasingly demand precise and high-quality paper recommendations. Among various recommendation methodologies, graph-based approaches have garnered attention by effectively exploiting the structural characteristics inherent in scholarly networks. However, these methods often overlook the asymmetric academic influence that is prevalent in scholarly networks when learning graph representations. To address this limitation, this study proposes the Mutual-Influence-Aware Recommendation (MIARec) model, which employs a gravity-based approach to measure the mutual academic influence between scholars and incorporates this influence into the feature aggregation process during message propagation in graph representation learning. Additionally, the model utilizes a multi-channel aggregation method to capture both individual embeddings of distinct single relational sub-networks and their interdependent embeddings, thereby enabling a more comprehensive understanding of the heterogeneous scholarly network. Extensive experiments conducted on real-world datasets demonstrate that the MIARec model outperforms baseline models across three primary evaluation metrics, indicating its effectiveness in scientific paper recommendation tasks.

cs.IR↗

arXiv2025

Beyond Pairwise Connections: Extracting High-Order Functional Brain Network Structures under Global Constraints

Functional brain network (FBN) modeling often relies on local pairwise interactions, whose limitation in capturing high-order dependencies is theoretically analyzed in this paper. Meanwhile, the computational burden and heuristic nature of current hypergraph modeling approaches hinder end-to-end learning of FBN structures directly from data distributions. To address this, we propose to extract high-order FBN structures under global constraints, and implement this as a Global Constraints oriented Multi-resolution (GCM) FBN structure learning framework. It incorporates 4 types of global constraint (signal synchronization, subject identity, expected edge numbers, and data labels) to enable learning FBN structures for 4 distinct levels (sample/subject/group/project) of modeling resolution. Experimental results demonstrate that GCM achieves up to a 30.6% improvement in relative accuracy and a 96.3% reduction in computational time across 5 datasets and 2 task settings, compared to 9 baselines and 10 state-of-the-art methods. Extensive experiments validate the contributions of individual components and highlight the interpretability of GCM. This work offers a novel perspective on FBN structure learning and provides a foundation for interdisciplinary applications in cognitive neuroscience. Code is publicly available on https://github.com/lzhan94swu/GCM.

cs.LG↗

arXiv2025

Generative Data Augmentation in Graph Contrastive Learning for Recommendation

Recommendation systems have become indispensable in various online platforms, from e-commerce to streaming services. A fundamental challenge in this domain is learning effective embeddings from sparse user-item interactions. While contrastive learning has recently emerged as a promising solution to this issue, generating augmented views for contrastive learning through most existing random data augmentation methods often leads to the alteration of original semantic information. In this paper, we propose a novel framework, GDA4Rec (Generative Data Augmentation in graph contrastive learning for Recommendation) to generate high-quality augmented views and provide robust self-supervised signals. Specifically, we employ a noise generation module that leverages deep generative models to approximate the distribution of original data for data augmentation. Additionally, GDA4Rec further extracts an item complement matrix to characterize the latent correlations between items and provide additional self-supervised signals. Lastly, a joint objective that integrates recommendation, data augmentation and contrastive learning is used to enforce the model to learn more effective and informative embeddings. Extensive experiments are conducted on three public datasets to demonstrate the superiority of the model. The code is available at: https://github.com/MrYansong/GDA4Rec.

cs.IR↗

arXiv2025

Deep learning framework for predicting stochastic take-off and die-out of early spreading

Large-scale outbreaks of epidemics, misinformation, or other harmful contagions pose significant threats to human society, yet the fundamental question of whether an emerging outbreak will escalate into a major epidemic or naturally die out remains largely unaddressed. This problem is challenging, partially due to inadequate data during the early stages of outbreaks and also because established models focus on average behaviors of large epidemics rather than the stochastic nature of small transmission chains. Here, we introduce the first systematic framework for forecasting whether initial transmission events will amplify into major outbreaks or fade into extinction during early stages, when intervention strategies can still be effectively implemented. Using extensive data from stochastic spreading models, we developed a deep learning framework that predicts early-stage spreading outcomes in real-time. Validation across Erdős-Rényi and Barabási-Albert networks with varying infectivity levels shows our method accurately forecasts stochastic spreading events well before potential outbreaks, demonstrating robust performance across different network structures and infectivity scenarios.To address the challenge of sparse data during early outbreak stages, we further propose a pretrain-finetune framework that leverages diverse simulation data for pretraining and adapts to specific scenarios through targeted fine-tuning. The pretrain-finetune framework consistently outperforms baseline models, achieving superior performance even when trained on limited scenario-specific data. To our knowledge, this work presents the first framework for predicting stochastic take-off versus die-out. This framework provides valuable insights for epidemic preparedness and public health decision-making, enabling more informed early intervention strategies.

cs.SI↗

arXiv2025

Apple Intelligence Foundation Language Models: Tech Report 2025

We introduce two multilingual, multimodal foundation language models that power Apple Intelligence features across Apple devices and services: i a 3B-parameter on-device model optimized for Apple silicon through architectural innovations such as KV-cache sharing and 2-bit quantization-aware training; and ii a scalable server model built on a novel Parallel-Track Mixture-of-Experts PT-MoE transformer that combines track parallelism, mixture-of-experts sparse computation, and interleaved global-local attention to deliver high quality with competitive cost on Apple's Private Cloud Compute platform. Both models are trained on large-scale multilingual and multimodal datasets sourced via responsible web crawling, licensed corpora, and high-quality synthetic data, then further refined with supervised fine-tuning and reinforcement learning on a new asynchronous platform. The resulting models support several additional languages while understanding images and executing tool calls. In public benchmarks and human evaluations, both the server model and the on-device model match or surpass comparably sized open baselines. A new Swift-centric Foundation Models framework exposes guided generation, constrained tool calling, and LoRA adapter fine-tuning, allowing developers to integrate these capabilities with a few lines of code. The latest advancements in Apple Intelligence models are grounded in our Responsible AI approach with safeguards like content filtering and locale-specific evaluation, as well as our commitment to protecting our users' privacy with innovations like Private Cloud Compute.

cs.LG↗