PRO-LONG: Programmatic Memory Enables Long-Horizon Reasoning· 程序化记忆框架提升长时间任务处理能力
Long-horizon tasks require sustained perception, reasoning, and exploration, and are a persistent challenge for large language model (LLM) agents. This gap is reflected in their limited performance on continual learning benchmarks such as ARC-AGI-3, especially when models are evaluated out of the box. Various agent harnesses have been proposed to close this gap, and each commits to a strategy for handling long sequences of observations, i.e., what information to save from the environment and how to load it into model context, a choice we argue is particularly consequential. Existing methods for context management face a significant tradeoff, as preserving more information makes retrieving relevant details less tractable. We propose PRO-LONG, a minimal context management framework built aro
提出程序化记忆框架 PRO-LONG 以改善 LLM 在长时间任务中的表现。
- 核心方法
- PRO-LONG 通过一种最小化的情境管理框架,优化代理在长时间任务中的信息保存和加载策略,以平衡信息量和检索效率。
- 适合谁读
- 研究者、工程师
- 要解决的问题
- 长时间任务处理中,维持持续的感知、推理和探索能力对 LLM 代理来说是一个挑战,尤其在持续学习基准测试中表现不佳。
- 关键实验
- 未提供
- 主要贡献
- 提供了一种有效的情境管理方法,提升了 LLM 代理在长时间任务和持续学习任务中的性能。
- 意义与局限
- 意义在于为解决 LLM 代理在长时间任务中的局限性提供了新的思路,但可能需进一步实验验证其在更广泛任务中的有效性。