递归深度过深反而让大模型‘过度思考’，导致效率暴跌
Think, But Don't Overthink: Reproducing Recursive Language Models
- 将提示词外置于外部环境，通过递归调用扩展上下文处理能力
- 递归深度从1增至2时，推理耗时从3.6秒飙升至344.5秒，准确率下降
- 适合研究大模型推理机制与资源消耗平衡的开发者参考
本项目复现并拓展了张等人（2026）提出的递归语言模型（RLM）框架。该框架通过将提示词外置于外部REPL环境，使大语言模型（LLM）能够处理近似无限长度的上下文。原论文默认使用递归深度1，并建议未来探索更深的递归。本研究特别考察了递归深度的影响。在S-NIAH和OOLONG基准测试中，使用开源代理模型DeepSeek v3.2和Kimi K2，评估了纯LLM、RLM（深度=1）和RLM（深度=2）的表现。结果发现：深度为1的RLM显著提升复杂推理任务准确率；但当递归深度增至2或在简单检索任务中使用RLM时，性能反而下降，执行时间呈指数级增长（如从3.6秒增至344.5秒），令牌成本大幅上升。代码与数据已公开于GitHub。
原文摘要 · Abstract (English)
This project reproduces and extends the recently proposed ``Recursive Language Models'' (RLMs) framework by Zhang et al. (2026). This framework enables Large Language Models (LLMs) to process near-infinite contexts by offloading the prompt into an external REPL environment. While the original paper relies on a default recursion depth of 1 and suggests deeper recursion as a future direction, this study specifically investigates the impact of scaling the recursion depth. Using state-of-the-art open-source agentic models (DeepSeek v3.2 and Kimi K2), I evaluated pure LLM, RLM (depth=1), and RLM (depth=2) on the S-NIAH and OOLONG benchmarks. The findings reveal a compelling phenomenon: Deeper recursion causes models to ``overthink''. While depth-1 RLMs effectively boost accuracy on complex reasoning tasks, applying deeper recursion (depth=2) or using RLMs on simple retrieval tasks paradoxically degrades performance and exponentially inflates execution time (e.g., from 3.6s to 344.5s) and token costs. Code and data are available at: https://github.com/drbillwang/rlm-reproduction
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。