4 papers
WISE-Flow: Workflow-Induced Structured Experience for Self-Evolving Conversational Service Agents
Yuqing Zhou, Zhuoer Wang, Jie Yuan +4
Large language model (LLM)-based agents are widely deployed in user-facing services but remain error-prone in new tasks, tend to repeat the same failure patterns, and show substant…
EDIT: Early Diffusion Inference Termination for dLLMs Based on Dynamics of Training Gradients
He-Yen Hsieh, Hong Wang, H. T. Kung
Diffusion-based large language models (dLLMs) refine token generations through iterative denoising, but answers often stabilize before all steps complete. We propose EDIT (Early Di…
GAPO: Robust Advantage Estimation for Real-World Code LLMs
Jianqing Zhang, Zhezheng Hao, Wei Xia +7
Reinforcement learning (RL) is widely used for post-training large language models (LLMs) in code editing, where group-relative methods, such as GRPO, are popular due to their crit…
Can We Trust the AI Pair Programmer? Copilot for API Misuse Detection and Correction
Saikat Mondal, Chanchal K. Roy, Hong Wang +2
API misuse introduces security vulnerabilities, system failures, and increases maintenance costs, all of which remain critical challenges in software development. Existing detectio…