AgentOCR: 通过光学自压缩重新想象代理历史
AgentOCR: Reimagining Agent History via Optical Self-Compression
AI总结：
AgentOCR通过光学自压缩技术，将代理历史转换为紧凑图像，从而在保持性能的同时显著减少token消耗和内存使用。
摘要：
近年来，大型语言模型（LLMs）的进步使得通过多轮交互轨迹进行强化学习（RL）训练的代理系统成为可能，但实际部署受到快速增长的文本历史所限制，因为这些历史会增加token预算和内存使用。我们引入了AgentOCR框架，该框架利用视觉token的优越信息密度，将累积的观察-动作历史表示为紧凑的渲染图像。为了使多轮回滚可扩展，AgentOCR提出了一种分段光学缓存机制。通过将历史分解为可哈希的分段并维护一个视觉缓存，该机制消除了冗余的重新渲染。除了固定渲染外，AgentOCR还引入了代理自压缩，其中代理主动发出压缩率，并通过压缩感知奖励进行训练，以自适应地平衡任务成功和token效率。我们在具有挑战性的代理基准测试中进行了广泛的实验，ALFWorld和基于搜索的问答。令人印象深刻的是，结果表明，AgentOCR在保持超过95%的基于文本的代理性能的同时，显著减少了token消耗（>50%），产生了一致的token和内存效率。我们的进一步分析验证了从分段光学缓存中获得的20倍渲染速度提升以及自压缩的有效战略平衡。
英文摘要：
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 exploits visual tokens' superior information density by representing the accumulated observation-action history as a compact rendered image. To make multi-turn rollouts scalable, AgentOCR proposes segment optical caching. By decomposing history into hashable segments and maintaining a visual cache, this mechanism eliminates redundant re-rendering. Beyond fixed rendering, AgentOCR introduces agentic self-compression, where the agent actively emits a compression rate and is trained with compression-aware reward to adaptively balance task success and token efficiency. We conduct extensive experiments on challenging agentic benchmarks, ALFWorld and search-based QA. Remarkably, AgentOCR preserves over 95\% of text-based agent performance while substantially reducing token consumption (>50\%), yielding consistent token and memory efficiency. Further analysis validates a 20$\times$ rendering speedup from optical caching and effective self-compression balancing. Our code is available at https://github.com/langfengQ/AgentOCR.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
