Zilong Wang

Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfaces

Memorize or Generalize? Evaluating LLM Code Generation with Code Rewriting

A Tale of LLMs and Induced Small Proxies: Scalable Agents for Knowledge Mining

Training Language Models to Generate Quality Code with Program Analysis Feedback

RRO: LLM Agent Optimization Through Rising Reward Trajectories

Speculative RAG: Enhancing Retrieval Augmented Generation through Drafting