研究显示：定期“遗忘”能让 LLM 智能体的自动探索走得更远
Scaling Discovery Requires Forgetting
几个 LLM 智能体做自动科研会越跑越僵，这篇论文发现定期清空记忆、只留最优解重启，能多挖出最多 39% 的改进。
一篇 arXiv 论文（arXiv:2610.20563）研究 LLM 智能体做自动科研的三种扩展方式：单智能体延长迭代、多智能体并行、并行加通信。实验发现三种方式都会随历史积累而陷入平台期，智能体被过去的尝试锚定，改动越来越少、越来越小。论文提出周期性清空搜索状态、只保留当前最优候选再重启，在启发式编程、内核优化和模型训练四类任务上，比并行智能体最高提升 15%，比单个智能体最高提升 39%。
Scaling Discovery Requires Forgetting
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 the candidate history, and restarting from the best candidate found so far, unblocks the search and pushes the same budget further. We implement forgetting in a team of agents with communication and evaluate it on four tasks spanning heuristic programming, kernel optimization and model training. The results show up to 15% improvement compared to parallel agents, and up to 39% compared to a single agent. Our results suggest that, for open-ended discovery, what an agent is allowed to remember matters as much as how much compute it is given.
