9 papers
ForkPilot: Self-Evolving Policy for Retrospective Search in Long-Horizon Agents
Xinyue Zeng, Shivam Shandilya, Guilherme Potje +6
Interactive language-model agents increasingly solve complex tasks through long-horizon, multi-call reasoning, where errors in beliefs or actions can compound across tool interacti…
How Can Rhetoric Reward-Hack AI Reviewers? Dissecting Rhetorical Sensitivity in AI-Based Peer Review
Ming Li, Chenguang Wang, Xirui Li +5
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…
Cognitive Episodes in LLM Reasoning Traces Enable Interpretable Human Item Difficulty Prediction
Chenguang Wang, Ming Li, Xinyue Zeng +4
Predicting human item difficulty is central to educational assessment, where reliable estimates support fairness and effective test construction. Existing methods often depend on c…
HalluGuard: Demystifying Data-Driven and Reasoning-Driven Hallucinations in LLMs
Xinyue Zeng, Junhong Lin, Yujun Yan +4
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 typi…
DiagnoLLM: A Hybrid Bayesian Neural Language Framework for Interpretable Disease Diagnosis
Bowen Xu, Xinyue Zeng, Jiazhen Hu +2
Building trustworthy clinical AI systems requires not only accurate predictions but also transparent, biologically grounded explanations. We present \texttt{DiagnoLLM}, a hybrid fr…
DISPROTBENCH: Uncovering the Functional Limits of Protein Structure Prediction Models in Intrinsically Disordered Regions
Xinyue Zeng, Tuo Wang, Adithya Kulkarni +6
Intrinsically disordered regions (IDRs) play central roles in cellular function, yet remain poorly evaluated by existing protein structure prediction benchmarks. Current evaluation…
LENSLLM: Unveiling Fine-Tuning Dynamics for LLM Selection
Xinyue Zeng, Haohui Wang, Junhong Lin +3
The proliferation of open-sourced Large Language Models (LLMs) and diverse downstream tasks necessitates efficient model selection, given the impracticality of fine-tuning all cand…
Scientific Hypothesis Generation and Validation: Methods, Datasets, and Future Directions
Adithya Kulkarni, Fatimah Alotaibi, Xinyue Zeng +7
Large Language Models (LLMs) are transforming scientific hypothesis generation and validation by enabling information synthesis, latent relationship discovery, and reasoning augmen…
Plan and Budget: Effective and Efficient Test-Time Scaling on Reasoning Large Language Models
Junhong Lin, Xinyue Zeng, Jie Zhu +4
Large Language Models (LLMs) have achieved remarkable success in complex reasoning tasks, but their inference remains computationally inefficient. We observe a common failure mode…