Scaling Discovery Requires Forgetting· 规模化发现需要忘记
Automated research with LLM agents has shown promising results in various domains. Through iterative refinement and scored verifiers, LLM agents are capable of discovering new ideas. There are different ways to scale automated research: let one agent iterate for longer, run several agents in parallel, or further let the parallel agents communicate and build on one another's candidates. However, we find that all three strategies eventually plateau as the agents accumulate history. Each agent becomes anchored to its own past attempts, making fewer and smaller edits. Beyond this point, additional budget brings little further improvement. In this paper, we show that the key to scaling discovery further is to forget. Periodically discarding the search state, including every agent's session and
通过定期遗忘以避免历史累积导致的探索停滞，从而进一步规模化自动化研究。
- 核心方法
- 提出通过定期清除代理的搜索状态（包括会话历史）来打破对过去尝试的依赖，使代理能够进行更大胆的探索。
- 适合谁读
- 研究者、工程师
- 要解决的问题
- 自动化研究的三种扩展策略（单代理长时间迭代、多代理并行运行、代理间通信）最终会因为历史累积而达到上限，无法进一步提高研究效率。
- 关键实验
- 未提供
- 主要贡献
- 证明了定期遗忘是克服自动化研究扩展瓶颈的关键方法，能够显著提高新想法的发现效率。
- 意义与局限
- 为自动化研究的规模化提供了新的思路，但需要进一步实验验证其在不同领域和任务中的适用性。