两大型模型互答时会陷入重复循环，自发形成收敛现象。
Convergence of Outputs When Two Large Language Models Interact in a Multi-Agentic Setup
- 两模型交替生成回应，无外部干预下逐步趋同。
- 多数对话从连贯转为重复，短语反复出现且输出趋同。
- 即使模型独立训练，仍会在多轮互动中自发收敛。
本文研究了两个大语言模型在无外界输入的多智能体环境下持续互答的现象。实验以一句简短种子句启动，随后两模型轮流阅读对方输出并生成响应，共进行固定轮次。使用Mistral Nemo Base 2407和Llama 2 13B hf作为模型。观察发现，多数对话初期保持连贯，但随后逐渐陷入重复。在多次运行中，一个短语反复出现并持续传递。一旦重复开始，两模型倾向于生成相似内容，不再引入新方向，形成文本循环。这种行为被定义为一种收敛。该现象即使在模型规模大、独立训练且无提示指令的情况下依然发生。我们通过词汇与嵌入度量方法，分析对话偏离初始种子的程度及两模型输出相似性的变化趋势。
原文摘要 · Abstract (English)
In this work, we report what happens when two large language models respond to each other for many turns without any outside input in a multi-agent setup. The setup begins with a short seed sentence. After that, each model reads the other's output and generates a response. This continues for a fixed number of steps. We used Mistral Nemo Base 2407 and Llama 2 13B hf. We observed that most conversations start coherently but later fall into repetition. In many runs, a short phrase appears and repeats across turns. Once repetition begins, both models tend to produce similar output rather than introducing a new direction in the conversation. This leads to a loop where the same or similar text is produced repeatedly. We describe this behavior as a form of convergence. It occurs even though the models are large, trained separately, and not given any prompt instructions. To study this behavior, we apply lexical and embedding-based metrics to measure how far the conversation drifts from the initial seed and how similar the outputs of the two models becomes as the conversation progresses.
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。