6 papers
Scheduling the Unschedulable: Taming Black-Box LLM Inference at Scale
Renzhong Yuan, Yijun Zeng, Xiaosong Gao +3
When output token counts can be predicted at submission time (Gan et al., 2026), client-side scheduling against a black-box LLM API becomes semi-clairvoyant: decisions condition on…
SkillRL: Evolving Agents via Recursive Skill-Augmented Reinforcement Learning
Peng Xia, Jianwen Chen, Hanyang Wang +10
Large Language Model (LLM) agents have shown stunning results in complex tasks, yet they often operate in isolation, failing to learn from past experiences. Existing memory-based m…
Text2Grad: Reinforcement Learning from Natural Language Feedback
Hanyang Wang, Lu Wang, Chaoyun Zhang +5
Traditional RLHF optimizes language models with coarse, scalar rewards that mask the fine-grained reasons behind success or failure, leading to slow and opaque learning. Recent wor…
Fine-Tuning without Performance Degradation
Han Wang, Adam White, Martha White
Fine-tuning policies learned offline remains a major challenge in application domains. Monotonic performance improvement during \emph{fine-tuning} is often challenging, as agents t…
Fat-to-Thin Policy Optimization: Offline RL with Sparse Policies
Lingwei Zhu, Han Wang, Yukie Nagai
Sparse continuous policies are distributions that can choose some actions at random yet keep strictly zero probability for the other actions, which are radically different from the…
q-exponential family for policy optimization
Lingwei Zhu, Haseeb Shah, Han Wang +2
Policy optimization methods benefit from a simple and tractable policy parametrization, usually the Gaussian for continuous action spaces. In this paper, we consider a broader poli…