Searcharxiv⌕ Search

arXiv subjects

Junjie Wang

Publications and source records attributed to Junjie Wang.

At least 73 records · Page 4Linked to original sources
arXiv2026

Think-with-Rubrics: From External Evaluator to Internal Reasoning Guidance

Rubrics have been extensively utilized for evaluating unverifiable, open-ended tasks, with recent research incorporating them into reward systems for reinforcement learning. However, existing frameworks typically treat rubrics only as external evaluator disjointed from the policy's primary reasoning trace. Such design confines rubrics to post-hoc measurement, leaving them unable to actively guide the model's generation process. In this work, we introduce Think-with-Rubrics, a novel paradigm for instruction following tasks. Think-with-Rubrics integrates rubric generation into the reasoning context, transforming the rubric from an independent artifact into an internal guidance of LLM's generation. During training, LLM sequentially generates a rubric followed by a response, while a trained rubric verifier provides joint supervision by evaluating the consistency between the answer and the self-generated / golden rubrics. Experiments across multiple benchmarks demonstrate that Think-with-Rubrics consistently outperforms the Rubric-as-Reward baseline supervised by golden rubrics by an average of 3.87 points. We have also discussed the mechanism by which Think-with-Rubrics enhances model performance. Experimental results demonstrate that supervision from golden rubrics and self-generated rubrics enhances the performance of Think-with-Rubrics by improving the quality of self-generated rubrics and increasing the internal consistency of responses respectively.

cs.CL↗
arXiv2026

ProFit: Leveraging High-Value Signals in SFT via Probability-Guided Token Selection

Supervised fine-tuning (SFT) is a fundamental post-training strategy to align Large Language Models (LLMs) with human intent. However, traditional SFT often ignores the one-to-many nature of language by forcing alignment with a single reference answer, leading to the model overfitting to non-core expressions. Although our empirical analysis suggests that introducing multiple reference answers can mitigate this issue, the prohibitive data and computational costs necessitate a strategic shift: prioritizing the mitigation of single-reference overfitting over the costly pursuit of answer diversity. To achieve this, we reveal the intrinsic connection between token probability and semantic importance: high-probability tokens carry the core logical framework, while low-probability tokens are mostly replaceable expressions. Based on this insight, we propose ProFit, which selectively masks low-probability tokens to prevent surface-level overfitting. Extensive experiments confirm that ProFit consistently outperforms traditional SFT baselines on general reasoning and mathematical benchmarks.

cs.CL↗
arXiv2026

GroupRank: A Groupwise Paradigm for Effective and Efficient Passage Reranking with LLMs

Large Language Models (LLMs) have emerged as powerful tools for passage reranking in information retrieval, leveraging their superior reasoning capabilities to address the limitations of conventional models on complex queries. However, current LLM-based reranking paradigms are fundamentally constrained by an efficiency-accuracy trade-off: (1) pointwise methods are efficient but ignore inter-document comparison, yielding suboptimal accuracy; (2) listwise methods capture global context but suffer from context-window constraints and prohibitive inference latency. To address these issues, we propose GroupRank, a novel paradigm that balances flexibility and context awareness. To unlock the full potential of groupwise reranking, we propose an answer-free data synthesis pipeline that fuses local pointwise signals with global listwise rankings. These samples facilitate supervised fine-tuning and reinforcement learning, with the latter guided by a specialized group-ranking reward comprising ranking-utility and group-alignment. These complementary components synergistically optimize document ordering and score calibration to reflect intrinsic query-document relevance. Experimental results show GroupRank achieves a state-of-the-art 65.2 NDCG@10 on BRIGHT and surpasses baselines by 2.1 points on R2MED, while delivering a 6.4$\times$ inference speedup.

cs.IR↗
arXiv2026

Where Did It Go Wrong? Capability-Oriented Failure Attribution for Vision-and-Language Navigation Agents

Embodied agents in safety-critical applications such as Vision-Language Navigation (VLN) rely on multiple interdependent capabilities (e.g., perception, memory, planning, decision), making failures difficult to localize and attribute. Existing testing methods are largely system-level and provide limited insight into which capability deficiencies cause task failures. We propose a capability-oriented testing approach that enables failure detection and attribution by combining (1) adaptive test case generation via seed selection and mutation, (2) capability oracles for identifying capability-specific errors, and (3) a feedback mechanism that attributes failures to capabilities and guides further test generation. Experiments show that our method discovers more failure cases and more accurately pinpoints capability-level deficiencies than state-of-the-art baselines, providing more interpretable and actionable guidance for improving embodied agents.

cs.MA↗
arXiv2026

Linear-Nonlinear Fusion Neural Operator for Partial Differential Equations

Neural operator learning directly constructs the mapping relationship from the equation parameter space to the solution space, enabling efficient direct inference in practical applications without the need for repeated solution of partial differential equations (PDEs) -- an advantage that is difficult to achieve with traditional numerical methods. In this work, we find that explicitly decoupling linear and nonlinear effects within such operator mappings leads to improved learning efficiency. This yields a novel network structure, namely the Linear-Nonlinear Fusion Neural Operator (LNF-NO), which models operator mappings via the multiplicative fusion of a linear component and a nonlinear component, thus achieving a lightweight and interpretable representation. This linear-nonlinear decoupling enables efficient capture of complex solution features at the operator level while maintaining stability and generality. LNF-NO naturally supports multiple functional inputs and is applicable to both regular grids and irregular geometries. Across a diverse suite of PDE operator-learning benchmarks, including nonlinear Poisson-Boltzmann equations and multi-physics coupled systems, LNF-NO is typically substantially faster to train than several representative neural operator baselines, while achieving comparable or improved accuracy across most tested cases. On the tested 3D Poisson-Boltzmann case, LNF-NO achieves strong accuracy while requiring substantially less training time than the three-dimensional Fourier Neural Operator and Transolver baselines.

cs.LG↗
arXiv2026

Seeing the Whole Elephant: A Benchmark for Failure Attribution in LLM-based Multi-Agent Systems

Failure attribution, i.e., identifying the responsible agent and decisive step of a failure, is particularly challenging in LLM-based multi-agent systems (MAS) due to their natural-language reasoning, nondeterministic outputs, and intricate interaction dynamics. A reliable benchmark is therefore essential to guide and evaluate attribution techniques. Yet existing benchmarks rely on partially observable traces that capture only agent outputs, omitting the inputs and context that developers actually use when debugging. We argue that failure attribution should be studied under full execution observability, aligning with real-world developer-facing scenarios where complete traces, rather than only outputs, are accessible for diagnosis. To this end, we introduce TraceElephant, a benchmark designed for failure attribution with full execution traces and reproducible environments. We then systematically evaluate failure attribution techniques across various configurations. Specifically, full traces improve attribution accuracy by up to 76\% over a partial-observation counterpart, confirming that missing inputs obscure many failure causes. TraceElephant provides a foundation for follow-up failure attribution research, promoting evaluation practices that reflect real-world debugging and supporting the development of more transparent MASs.

cs.MA↗
arXiv2026

Amorphous Nanoconfinement Enables Self-sustaining Sabatier Reaction at Ambient Conditions

The Sabatier reaction, the catalytic hydrogenation of CO2 into CH4, offers a cornerstone for carbon capture and utilization, and in-situ resource utilization during space exploration; however, it faces a fundamental thermodynamic-kinetic paradox: although highly exothermic, conventional catalysts still require continuous external heating to activate CO2 and maintain stable operation. Here we report an amorphous silica-embedded ruthenium catalyst that enables a long-term self-sustaining autothermal Sabatier reaction dispensing with external energy supply. Operating under ambient conditions, this system achieves a record-high CH4 yield of 0.50 mol gcat-1 h-1 with 100% selectivity, stable operation for over 2,000 hours, and a record-low catalyst bed temperature down to 100 oC. This exceptional self-sustaining behavior stems from the synergistic effect of the catalyst's ultralow effective thermal conductivity (0.27 W m-1 K-1), induced by amorphous nanoconfinement, and its superior intrinsic activity. This synergy generates localized hot spots at Ru sites while suppressing macroscopic heat loss. In situ measurements further reveal CH4 formation even at 54 oC and identify a *CO-mediated pathway for CO2 methanation. The reaction ignites readily with a lighter or focused sunlight and persists even under forced convection from an electric fan, demonstrating strong environmental tolerance. By removing the need for constant energy input, this "ignite-and-forget" system paves the way for decentralized Power-to-Gas systems and autonomous fuel production in resource-constrained environments like Mars.

cond-mat.mtrl-sci↗
arXiv2026

Interplay of Flat-band and Anderson localizations in disordered moire superlattices

Disorder in moire superlattices simultaneously degrades flat-band localization and induces Anderson localization, yet how these two regimes interact has remained unclear. Here, we introduce a combined framework linking localization-length scaling with differential probability density analysis to map localization transitions in partially disordered one-dimensional silicon moire lattices. It is found that flat bands confined within the interband gap keep their strong localization even as disorder grows. In contrast, flat bands intersecting dispersive bands exhibit rich behaviors: the low-frequency branch undergoes an inverse Anderson transition, while the high-frequency branch supports coexisting flat-band and Anderson localization at strong disorder. Our results deliver the direct evidence of competing localization mechanisms in disordered moire systems and offer guiding principles for engineering robust, nonideal moire photonic devices.

cond-mat.dis-nn↗
arXiv2026

Hybrid Architectures for Language Models: Systematic Analysis and Design Insights

Recent progress in large language models demonstrates that hybrid architectures--combining self-attention mechanisms with structured state space models like Mamba--can achieve a compelling balance between modeling quality and computational efficiency, particularly for long-context tasks. While these hybrid models show promising performance, systematic comparisons of hybridization strategies and analyses on the key factors behind their effectiveness have not been clearly shared to the community. In this work, we present a holistic evaluation of hybrid architectures based on inter-layer (sequential) or intra-layer (parallel) fusion. We comprehensively evaluate these designs across multiple dimensions: language modeling and downstream task performance, long-context capabilities, scaling analysis, and training and inference efficiency. By investigating the core characteristics of their computational primitive, we identify the most critical elements for each hybridization strategy and further propose optimal design recipes for hybrid models. Our comprehensive analysis provides practical guidance and valuable insights for developing hybrid language models, facilitating the optimization of architectural configurations.

cs.CL↗
arXiv2026

Fall into a Pit, Gain in a Wit: Cognitive-Guided Harmful Meme Detection via Misjudgment Risk Pattern Retrieval

Internet memes have emerged as a popular multimodal medium, yet they are increasingly weaponized to convey harmful opinions through subtle rhetorical devices like irony and metaphor. Existing detection approaches, including Multimodal Large Language Model (MLLM)-based techniques, struggle with these implicit expressions, leading to frequent misjudgments. This paper introduces PatMD, a novel approach that detects harmful memes by learning from and proactively mitigating these potential misjudgment risks. Our core idea is to move beyond superficial content-level matching and instead identify the underlying misjudgment risk patterns, proactively guiding the MLLMs to avoid known misjudgment pitfalls. We first construct a knowledge base where each meme is deconstructed into a misjudgment risk pattern explaining why it might be misjudged, either overlooking harmful undertones (false negative) or overinterpreting benign content (false positive). For a given target meme, PatMD retrieves relevant patterns and utilizes them to dynamically guide the MLLM's reasoning. Experiments on a benchmark of 6,626 memes across 5 harmful detection tasks show that PatMD outperforms state-of-the-art baselines, achieving an average of 8.30% improvement in F1-score and 7.71% improvement in accuracy, while exhibiting consistent robustness on unseen and adversarial memes.

cs.LG↗
arXiv2026

All Changes May Have Invariant Principles: Improving Ever-Shifting Harmful Meme Detection via Design Concept Reproduction

Harmful memes are ever-shifting in the Internet communities, which are difficult to analyze due to their type-shifting and temporal-evolving nature. Although these memes are shifting, we find that different memes may share invariant principles, i.e., the underlying design concept of malicious users, which can help us analyze why these memes are harmful. In this paper, we propose RepMD, an ever-shifting harmful meme detection method based on the design concept reproduction. We first refer to the attack tree to define the Design Concept Graph (DCG), which describes steps that people may take to design a harmful meme. Then, we derive the DCG from historical memes with design step reproduction and graph pruning. Finally, we use DCG to guide the Multimodal Large Language Model (MLLM) to detect harmful memes. The evaluation results show that RepMD achieves the highest accuracy with 81.1% and has slight accuracy decreases when generalized to type-shifting and temporal-evolving memes. Human evaluation shows that RepMD can improve the efficiency of human discovery on harmful memes, with 15$\sim$30 seconds per meme.

cs.CV↗
arXiv2026

NEPMaker: Active learning of neuroevolution machine learning potential for large cells

Machine learning potentials (MLPs) achieve near first-principles accuracy but often fail for atomic environments outside the training distribution. Active learning can mitigate this limitation; however, its application to large-scale simulations is hindered by the prohibitive cost of labeling entire configurations. Here, we develop a D-optimality-driven active learning framework for the neuroevolution potential (NEP) implemented within the GPUMD package, named NEPMaker. Extrapolative atomic environments are identified on-the-fly and embedded into locally periodic structures, where boundary atoms are optimized to remain close to the training distribution. This strategy enables large-scale simulations to directly contribute to dataset construction, significantly reducing extrapolation errors while improving model robustness and transferability. The proposed framework provides a scalable route for constructing reliable machine learning potentials in complex materials systems, including those involving defects, interfaces, and phase transitions.

physics.comp-ph↗
arXiv2026

One Shot Dominance: Knowledge Poisoning Attack on Retrieval-Augmented Generation Systems

Large Language Models (LLMs) enhanced with Retrieval-Augmented Generation (RAG) have shown improved performance in generating accurate responses. However, the dependence on external knowledge bases introduces potential security vulnerabilities, particularly when these knowledge bases are publicly accessible and modifiable. While previous studies have exposed knowledge poisoning risks in RAG systems, existing attack methods suffer from critical limitations: they either require injecting multiple poisoned documents (resulting in poor stealthiness) or can only function effectively on simplistic queries (limiting real-world applicability). This paper reveals a more realistic knowledge poisoning attack against RAG systems that achieves successful attacks by poisoning only a single document while remaining effective for complex multi-hop questions involving complex relationships between multiple elements. Our proposed AuthChain address three challenges to ensure the poisoned documents are reliably retrieved and trusted by the LLM, even against large knowledge bases and LLM's own knowledge. Extensive experiments across six popular LLMs demonstrate that AuthChain achieves significantly higher attack success rates while maintaining superior stealthiness against RAG defense mechanisms compared to state-of-the-art baselines.

cs.CR↗
arXiv2026

Know Thy Enemy: Securing LLMs Against Prompt Injection via Diverse Data Synthesis and Instruction-Level Chain-of-Thought Learning

Large language model (LLM)-integrated applications have become increasingly prevalent, yet face critical security vulnerabilities from prompt injection (PI) attacks. Defending against PI attacks faces two major issues: malicious instructions can be injected through diverse vectors, and injected instructions often lack clear semantic boundaries from the surrounding context, making them difficult to identify. To address these issues, we propose InstruCoT, a model enhancement method for PI defense that synthesizes diverse training data and employs instruction-level chain-of-thought fine-tuning, enabling LLMs to effectively identify and reject malicious instructions regardless of their source or position in the context. We evaluate InstruCoT across three critical dimensions: Behavior Deviation, Privacy Leakage, and Harmful Output. Experimental results across four LLMs demonstrate that InstruCoT significantly outperforms baselines in all dimensions while maintaining utility performance without degradation

cs.AI↗
arXiv2026

SpecQuant: Spectral Decomposition and Adaptive Truncation for Ultra-Low-Bit LLMs Quantization

The emergence of accurate open large language models (LLMs) has sparked a push for advanced quantization techniques to enable efficient deployment on end-user devices. In this paper, we revisit the challenge of extreme LLM compression -- targeting ultra-low-bit quantization for both activations and weights -- from a Fourier frequency domain perspective. We propose SpecQuant, a two-stage framework that tackles activation outliers and cross-channel variance. In the first stage, activation outliers are smoothed and transferred into the weight matrix to simplify downstream quantization. In the second stage, we apply channel-wise low-frequency Fourier truncation to suppress high-frequency components while preserving essential signal energy, improving quantization robustness. Our method builds on the principle that most of the weight energy is concentrated in low-frequency components, which can be retained with minimal impact on model accuracy. To enable runtime adaptability, we introduce a lightweight truncation module during inference that adjusts truncation thresholds based on channel characteristics. On LLaMA-3 8B, SpecQuant achieves 4-bit quantization for both weights and activations, narrowing the zero-shot accuracy gap to only 1.5% compared to full precision, while delivering 2 times faster inference and 3times lower memory usage. Code will be available at https://github.com/Kishon-zzx/SpecQuant.

cs.LG↗
arXiv2026

DIVER: A Multi-Stage Approach for Reasoning-intensive Information Retrieval

Retrieval-augmented generation has achieved strong performance on knowledge-intensive tasks where query-document relevance can be identified through direct lexical or semantic matches. However, many real-world queries involve abstract reasoning, analogical thinking, or multi-step inference, which existing retrievers often struggle to capture. To address this challenge, we present DIVER, a retrieval pipeline designed for reasoning-intensive information retrieval. It consists of four components. The document preprocessing stage enhances readability and preserves content by cleaning noisy texts and segmenting long documents. The query expansion stage leverages large language models to iteratively refine user queries with explicit reasoning and evidence from retrieved documents. The retrieval stage employs a model fine-tuned on synthetic data spanning medical and mathematical domains, along with hard negatives, enabling effective handling of reasoning-intensive queries. Finally, the reranking stage combines pointwise and listwise strategies to produce both fine-grained and globally consistent rankings. On the BRIGHT benchmark, DIVER achieves state-of-the-art nDCG@10 scores of 46.8 overall and 31.9 on original queries, consistently outperforming competitive reasoning-aware models. These results demonstrate the effectiveness of reasoning-aware retrieval strategies in complex real-world tasks.

cs.IR↗
arXiv2026

Efficient Reasoning with Balanced Thinking

Large Reasoning Models (LRMs) have shown remarkable reasoning capabilities, yet they often suffer from overthinking, expending redundant computational steps on simple problems, or underthinking, failing to explore sufficient reasoning paths despite inherent capabilities. These issues lead to inefficiencies and potential inaccuracies, limiting practical deployment in resource-constrained settings. Existing methods to mitigate overthinking, such as suppressing reflective keywords or adjusting reasoning length, may inadvertently induce underthinking, compromising accuracy. Therefore, we propose ReBalance, a training-free framework that achieves efficient reasoning with balanced thinking. ReBalance leverages confidence as a continuous indicator of reasoning dynamics, identifying overthinking through high confidence variance and underthinking via consistent overconfidence. By aggregating hidden states from a small-scale dataset into reasoning mode prototypes, we compute a steering vector to guide LRMs' reasoning trajectories. A dynamic control function modulates this vector's strength and direction based on real-time confidence, pruning redundancy during overthinking, and promoting exploration during underthinking. Extensive experiments conducted on four models ranging from 0.5B to 32B, and across nine benchmarks in math reasoning, general question answering, and coding tasks demonstrate that ReBalance effectively reduces output redundancy while improving accuracy, offering a general, training-free, and plug-and-play strategy for efficient and robust LRM deployment. Project page and code are available at https://rebalance-ai.github.io .

cs.AI↗
arXiv2026

Mapping generative AI use in the human brain: divergent neural, academic, and mental health profiles of functional versus socio emotional AI use

The widespread adoption of generative artificial intelligence conversational agents (AICAs) among university students constitutes a novel cognitive social environment whose impact on the maturing brain remains elusive. Combining surveys with high resolution structural MRI, we examined patterns of general, functional, and socio emotional AICA use, academic performance, mental health, and brain structural signatures in a comparatively large sample of 222 young individuals. Across computational anatomy, meta analytic network level, and behavioral decoding analyses, we observed use specific associations. Higher general and functional AICA use frequencies were linked to better academic outcomes (GPA), larger dorsolateral prefrontal and calcarine gray matter volume, and enhanced hippocampal network clustering and local efficiency. In contrast, more frequent socio emotional AICA use was associated with poorer mental health (depression, social anxiety) and lower volume of superior temporal and amygdalar regions central to social and affective processing. These findings indicate that the same class of AI tools exerts distinct effects depending on usage patterns and motivations, engaging prefrontal hippocampal systems that support cognition versus socio emotional systems that may track distress linked usage. These heterogeneities are crucial for designing environments that harness the educational benefits of AI while mitigating mental health risks.

q-bio.NC↗