7 papers
OTROPE: Optimal Transport-based Robust Off-policy Evaluation for Large Language Models
Liner Xiang, Wenbo Zhang, Hengrui Cai
Reliable evaluation of large language models (LLMs) is essential for their development and deployment, yet is often costly, risky, and difficult to perform safely online. We study…
UOPD: Uncertainty-Aware Intervention for On-Policy Distillation of Multi-Turn Agents
Wenbo Zhang, Pengcheng Xu, Weizhi Du +2
On-policy distillation (OPD) trains a student on its own rollouts using dense supervision from a teacher. In multi-turn environments, a mistake at a critical decision step can redi…
Reasoning Is Not Free: Robust Adaptive Cost-Efficient Routing for LLM-as-a-Judge
Wenbo Zhang, Lijinghua Zhang, Liner Xiang +1
Reasoning-capable large language models (LLMs) have recently been adopted as automated judges, but their benefits and costs in LLM-as-a-Judge settings remain unclear. Through contr…
Supervised Fine-Tuning versus Reinforcement Learning: A Study of Post-Training Methods for Large Language Models
Haitao Jiang, Wenbo Zhang, Jiarui Yao +3
Pre-trained Large Language Model (LLM) exhibits broad capabilities, yet, for specific tasks or domains their attainment of higher accuracy and more reliable reasoning generally dep…
Where to Intervene: Action Selection in Deep Reinforcement Learning
Wenbo Zhang, Hengrui Cai
Deep reinforcement learning (RL) has gained widespread adoption in recent years but faces significant challenges, particularly in unknown and complex environments. Among these, hig…
Beyond the Singular: Revealing the Value of Multiple Generations in Benchmark Evaluation
Wenbo Zhang, Hengrui Cai, Wenyu Chen
Large language models (LLMs) have demonstrated significant utility in real-world applications, exhibiting impressive capabilities in natural language processing and understanding.…
Recognizing Limits: Investigating Infeasibility in Large Language Models
Wenbo Zhang, Zihang Xu, Hengrui Cai
Large language models (LLMs) have shown remarkable performance in various tasks but often fail to handle queries that exceed their knowledge and capabilities, leading to incorrect…