作者位次 #4/6ICLR 2026Poster· Reinforcement Learning / Online
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。
Lang Feng · Fuchao Yang · Feng Chen · Xin Cheng · Haiyang Xu · Zhenglin Wan · Ming Yan · Bo An
Recent advances in large language models (LLMs) enable agentic systems trained with reinforcement learning (RL) over multi-turn interaction, but practical deployment is bottlenecked by rapidly growing textual histories that inflate token and memory costs. We introduce AgentOCR, a framework that e...