arXiv subjects

Xinyue Zeng

Publications and source records attributed to Xinyue Zeng.

8 recordsLinked to original sources

arXiv2026

How Can Rhetoric Reward-Hack AI Reviewers? Dissecting Rhetorical Sensitivity in AI-Based Peer Review

As large language models increasingly participate in scientific evaluation, we investigate a potential form of reward hacking: how rhetorical choices shape AI-review judgments when reported scientific content is preserved and how these effects vary across evaluation conditions. We construct a controlled corpus of 4,200 full-paper manuscripts derived from 120 anonymized ICLR 2026 submissions. Two LLM rewriters transform six rhetorical dimensions in opposing directions, and five LLM reviewers evaluate the resulting manuscripts under standard and strict protocols. We also test joint, recursive, and reviewer-guided rewriting. Our results show that rhetorical sensitivity is structured rather than uniform. Evidence framing and novelty stance produce the largest positive-negative contrasts in overall assessment, with scope framing forming a weaker second tier; the remaining dimensions have smaller or less stable effects. This hierarchy persists across human-assessed quality levels, but score movement depends strongly on the AI reviewer's original score: lower scores tend to rise, higher scores tend to fall, and directional contrasts are clearest in the middle ranges. More elaborate workflows do not reliably yield larger gains. Joint rewriting is strongly rewriter-dependent, reviewer guidance does not consistently outperform an unguided second pass, and repeated rewriting yields diminishing, configuration-dependent returns. Across conditions, the rewriter primarily determines the separation between opposing variants, whereas the reviewer determines the magnitude and sign of their score effects. Strict review lowers mean OA by 1.36 points without consistently changing rhetorical sensitivity. These findings identify when rhetorical presentation influences AI scientific review and motivate evaluation systems robust to content-preserving variation in scientific writing.

cs.CL↗

arXiv2026

Cognitive Episodes in LLM Reasoning Traces Enable Interpretable Human Item Difficulty Prediction

Predicting human item difficulty is central to educational assessment, where reliable estimates support fairness and effective test construction. Existing methods often depend on costly human calibration or item-level textual representations, providing limited evidence about the cognitive processes that make items difficult. We argue that difficulty should be viewed not only as a property of item text, but also as an observable consequence of the problem-solving burden an item induces. Large Reasoning Models (LRMs) offer scalable process evidence through reasoning traces, but such evidence must be structured to support interpretable modeling. To this end, we introduce Epi2Diff (Episode to Difficulty), a framework that maps LRM reasoning traces into cognitively grounded episode sequences. These episodes group trace segments into functional problem-solving states, enabling difficulty to be modeled through reasoning scale, effort allocation, and state transitions. Epi2Diff extracts compact episode-dynamic features and combines them with semantic item representations for human difficulty prediction. Experiments on four real-world human difficulty datasets show that Epi2Diff consistently outperforms strong baselines, including fine-tuned small language models, LLM in-context learning, and supervised LLM adaptation. On SAT-derived classification benchmarks, Epi2Diff achieves an 8.1% average relative gain over supervised LLM fine-tuning baselines. Further analyses show that harder items induce more effortful, iterative, and implementation-centered episode dynamics, rather than merely longer responses. These results demonstrate that cognitive episodes in LRM reasoning traces provide a predictive and interpretable process representation for human item difficulty, offering a new lens for educational measurement with reasoning models.

cs.CL↗

arXiv2026

HalluGuard: Demystifying Data-Driven and Reasoning-Driven Hallucinations in LLMs

The reliability of Large Language Models (LLMs) in high-stakes domains such as healthcare, law, and scientific discovery is often compromised by hallucinations. These failures typically stem from two sources: data-driven hallucinations and reasoning-driven hallucinations. However, existing detection methods usually address only one source and rely on task-specific heuristics, limiting their generalization to complex scenarios. To overcome these limitations, we introduce the Hallucination Risk Bound, a unified theoretical framework that formally decomposes hallucination risk into data-driven and reasoning-driven components, linked respectively to training-time mismatches and inference-time instabilities. This provides a principled foundation for analyzing how hallucinations emerge and evolve. Building on this foundation, we introduce HalluGuard, an NTK-based score that leverages the induced geometry and captured representations of the NTK to jointly identify data-driven and reasoning-driven hallucinations. We evaluate HalluGuard on 10 diverse benchmarks, 11 competitive baselines, and 9 popular LLM backbones, consistently achieving state-of-the-art performance in detecting diverse forms of LLM hallucinations. We open-source our proposed \model{} model at https://github.com/Susan571/HalluGuard-ICLR2026.

cs.LG↗

arXiv2025

DiagnoLLM: A Hybrid Bayesian Neural Language Framework for Interpretable Disease Diagnosis

Building trustworthy clinical AI systems requires not only accurate predictions but also transparent, biologically grounded explanations. We present \texttt{DiagnoLLM}, a hybrid framework that integrates Bayesian deconvolution, eQTL-guided deep learning, and LLM-based narrative generation for interpretable disease diagnosis. DiagnoLLM begins with GP-unmix, a Gaussian Process-based hierarchical model that infers cell-type-specific gene expression profiles from bulk and single-cell RNA-seq data while modeling biological uncertainty. These features, combined with regulatory priors from eQTL analysis, power a neural classifier that achieves high predictive performance in Alzheimer's Disease (AD) detection (88.0\% accuracy). To support human understanding and trust, we introduce an LLM-based reasoning module that translates model outputs into audience-specific diagnostic reports, grounded in clinical features, attribution signals, and domain knowledge. Human evaluations confirm that these reports are accurate, actionable, and appropriately tailored for both physicians and patients. Our findings show that LLMs, when deployed as post-hoc reasoners rather than end-to-end predictors, can serve as effective communicators within hybrid diagnostic pipelines.

cs.AI↗

arXiv2025

DISPROTBENCH: Uncovering the Functional Limits of Protein Structure Prediction Models in Intrinsically Disordered Regions

Intrinsically disordered regions (IDRs) play central roles in cellular function, yet remain poorly evaluated by existing protein structure prediction benchmarks. Current evaluations largely focus on well-folded domains, overlooking three fundamental challenges in realistic biological settings: the structural complexity of proteins, the resulting low availability of reliable ground truth, and prediction uncertainty that can propagate into high-risk downstream failures, such as in drug discovery, protein-protein interaction modeling, and functional annotation. We present DisProtBench, an IDR-centric benchmark that explicitly incorporates prediction uncertainty into the evaluation of protein structure prediction models (PSPMs). To address structural complexity and ground-truth scarcity, we curate and unify a large-scale, multi-modal dataset spanning disease-relevant IDRs, GPCR-ligand interactions, and multimeric protein complexes. To assess predictive uncertainty, we introduce Functional Uncertainty Sensitivity (FUS), a novel prediction uncertainty-stratified metric that quantifies downstream task performance under prediction uncertainty. Using this benchmark, we conduct a systematic evaluation of state-of-the-art PSPMs and reveal clear, task-dependent failure modes. Protein-protein interaction prediction degrades sharply in IDRs, while structure-based drug discovery remains comparatively robust. These effects are largely invisible to standard global accuracy metrics, which overestimate functional reliability under prediction uncertainty. We have open-sourced our benchmark and the codebase at https://github.com/Susan571/DisProtBench.

q-bio.BM↗

arXiv2025

Plan and Budget: Effective and Efficient Test-Time Scaling on Reasoning Large Language Models

Large Language Models (LLMs) have achieved remarkable success in complex reasoning tasks, but their inference remains computationally inefficient. We observe a common failure mode in many prevalent LLMs, overthinking, where models generate verbose and tangential reasoning traces even for simple queries. Recent work has tried to mitigate this by enforcing fixed token budgets, however, this can lead to underthinking, especially on harder problems. Through empirical analysis, we identify that this inefficiency often stems from unclear problem-solving strategies. To formalize this, we develop a theoretical model, BAM (Budget Allocation Model), which models reasoning as a sequence of sub-questions with varying uncertainty, and introduce the E3 metric to capture the trade-off between correctness and computation efficiency. Building on theoretical results from BAM, we propose Plan-and-Budget, a model-agnostic, test-time framework that decomposes complex queries into sub-questions and allocates token budgets based on estimated complexity using adaptive scheduling. Plan-and-Budget improves reasoning efficiency across a range of tasks and models, achieving up to 70% accuracy gains, 39% token reduction, and 193.8% improvement in E3. Notably, it improves the efficiency of a smaller model (DS-Qwen-32B) to match the efficiency of a larger model (DS-LLaMA-70B), demonstrating Plan-and-Budget's ability to close performance gaps without retraining. Our code is available at https://github.com/junhongmit/P-and-B.

cs.LG↗

arXiv2025

Scientific Hypothesis Generation and Validation: Methods, Datasets, and Future Directions

Large Language Models (LLMs) are transforming scientific hypothesis generation and validation by enabling information synthesis, latent relationship discovery, and reasoning augmentation. This survey provides a structured overview of LLM-driven approaches, including symbolic frameworks, generative models, hybrid systems, and multi-agent architectures. We examine techniques such as retrieval-augmented generation, knowledge-graph completion, simulation, causal inference, and tool-assisted reasoning, highlighting trade-offs in interpretability, novelty, and domain alignment. We contrast early symbolic discovery systems (e.g., BACON, KEKADA) with modern LLM pipelines that leverage in-context learning and domain adaptation via fine-tuning, retrieval, and symbolic grounding. For validation, we review simulation, human-AI collaboration, causal modeling, and uncertainty quantification, emphasizing iterative assessment in open-world contexts. The survey maps datasets across biomedicine, materials science, environmental science, and social science, introducing new resources like AHTech and CSKG-600. Finally, we outline a roadmap emphasizing novelty-aware generation, multimodal-symbolic integration, human-in-the-loop systems, and ethical safeguards, positioning LLMs as agents for principled, scalable scientific discovery.

cs.CL↗

arXiv2025

LENSLLM: Unveiling Fine-Tuning Dynamics for LLM Selection

The proliferation of open-sourced Large Language Models (LLMs) and diverse downstream tasks necessitates efficient model selection, given the impracticality of fine-tuning all candidates due to computational constraints. Despite the recent advances in LLM selection, a fundamental research question largely remains nascent: how can we model the dynamic behaviors of LLMs during fine-tuning, thereby enhancing our understanding of their generalization performance across diverse downstream tasks? In this work, we propose a novel theoretical framework that provides a proper lens to assess the generalization capabilities of LLMs, thereby enabling accurate and efficient LLM selection for downstream applications. In particular, we first derive a PAC-Bayesian Generalization Bound that unveils fine-tuning dynamics of LLMs and then introduce LENSLLM, a Neural Tangent Kernel (NTK)-based Rectified Scaling Model that enables accurate performance predictions across diverse tasks while maintaining computational efficiency. Extensive empirical results on 3 large-scale benchmarks demonstrate that our model achieves up to 91.1% accuracy and reduces up to 88.5% computational cost in LLM selection, outperforming 5 state-of-the-art methods. We open-source our proposed LENSLLM model and corresponding results at LensLLM.io.

cs.LG↗