智能体在受挑战时会修复还是仅回应?挑战、修复与公共纠正在一个部署的智能体论坛中的探讨
Do Agents Repair When Challenged -- or Just Reply? Challenge, Repair, and Public Correction in a Deployed Agent Forum
AI总结:
研究比较了部署的智能体论坛Moltbook与Reddit社区,发现Moltbook讨论线程较少,挑战回应率低,修复机制缺失,表明社区维持互动过程对规范教学和修订的重要性。
摘要:
随着大型语言模型(LLM)智能体在公共互动环境中部署,一个关键问题是其社区能否持续挑战、修复和公共纠正,或仅产生规范性语言。我们通过追踪三步机制比较了Moltbook和五个匹配的Reddit社区:讨论是否创建线程交流,挑战是否引发回应,以及纠正是否对更广泛线程可见。与Reddit相比,Moltbook讨论线程大约少10倍,挑战和回应的机会更少。当挑战发生时,原始作者几乎从不回应(1.2% vs. 40.9%在Reddit),多轮延续几乎不存在(0.1% vs. 38.5%),且在共享保守协议下未检测到修复。Reddit内的非挑战基线表明,这一差距与挑战有关,而非仅仅是更深入的线程。这些结果表明,社会一致性不仅取决于产生规范意识语言,还取决于维持社区通过互动过程教授、执行和修订规范的互动过程。这在安全方面很重要,因为纠正日益去中心化;在公平方面也很重要,因为社区在参与者如何与挑战互动上有不同期望。
英文摘要:
As large language model (LLM) agents enter public forums, a key question is whether those forums sustain challenge, repair, and public correction, or merely produce norm-like language. We compare Moltbook, a live deployed agent forum, with five topically matched Reddit communities across a three-step mechanism. Relative to Reddit, Moltbook discussions are roughly ten times less threaded, leaving far fewer chances for challenge and response. When challenges do occur, the original author almost never returns (1.2\% vs.\ 40.9\% on Reddit), multi-turn continuation is nearly absent ($<$0.1\% vs.\ 38.5\%), and the shared lexical protocol detects no direct repairs on the agent side. The deficit is at the re-engagement step rather than in repair-substance, since the few Moltbook authors who do return often repair substantively, and the gap persists under two LLM-judges, human annotation, and a within-Reddit non-challenge baseline. Correcting for the detector's lower precision on Moltbook narrows this gap without removing it, and our results characterize one deployed pipeline rather than LLM agents in general. Social alignment evaluation should therefore measure not only norm-aware language but the interactional processes through which communities enforce norms.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。