# 发现协作流水线：面向序列社会困境的自主研究

- 来源：HuggingFace Daily Papers（社区热门论文）
- 发布时间：2026-05-28 08:00
- AIHOT 分数：55
- AIHOT 链接：https://aihot.virxact.com/items/cmpqjlfr105jzslnom85e5dfc
- 原文链接：https://arxiv.org/abs/2605.30003

## AI 摘要

本研究构建了一个两层自主研究系统：外层AI智能体自主重新设计内层用于多智能体序列社会困境（SSDs）的大语言模型策略合成流水线。在Cleanup和Gathering两个游戏、两种福利目标（功利效率与最大化最小）下，该系统在性能上可靠地超越手工设计的基准，显著降低运行方差，并优于仅优化提示词的方法。研究发现，所发现的流水线具有目标依赖性：仅在最大化最小目标下，系统才会向合成器流水线注入显式的公平机制，这种机制在其自身的系统提示词和所有面向效率优化的流水线中均不存在，支持了信息设计理论的观点。

## 正文

We study two-level autoresearch for cooperation: an outer-loop AI agent autonomously redesigns the inner-loop pipeline of an LLM policy-synthesis system for multi-agent Sequential Social Dilemmas (SSDs). A researcher agent R (run as a coding agent) reads the inner-loop source code, edits system prompts, feedback functions, helper libraries, and iteration logic, runs evaluations, and decides what to keep, following the autoresearch paradigm. Across two games (Cleanup and Gathering), two policy-synthesizer LLMs, and two welfare objectives (utilitarian efficiency and Rawlsian maximin), the researcher reliably exceeds hand-designed baselines, sharply tightens run-to-run variance, and outperforms prompt-only optimization. The discovered pipelines are objective-dependent: only under maximin does the researcher inject an explicit fairness mechanism into synthesizer pipelines, a class of mechanism that is absent from its own objective-agnostic system prompt and from every efficiency-optimized pipeline. This supports an information-design reading in which the researcher chooses what to reveal to the boundedly rational synthesizer as a function of the welfare objective. Code at https://github.com/vicgalle/autoresearch-social-dilemmas.