作者位次 #2/5ICML 2026Poster· Social Aspects / Security
Training Language Model Agents to Find Vulnerabilities with CTF-Dojo
Terry Yue Zhuo · Dingmin Wang · Hantian Ding · Varun Kumar · Zijian Wang
首次获取该候选人的完整档案及 StartLux 评估共 20 积分；每位用户单独解锁，之后查看免费。主动刷新 20 积分。
实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。
Terry Yue Zhuo · Dingmin Wang · Hantian Ding · Varun Kumar · Zijian Wang
Myeongsoo Kim · Chao-Chun Hsu · Dingmin Wang · Shweta Garg · Varun Kumar · Murali Krishna Ramanathan
LLM-based code agents treat repositories as unstructured text, applying edits through brittle string matching that frequently fails due to formatting drift or ambiguous patterns. We propose reframing the codebase as a structured action space where agents operate on named AST entities rather than...