文献类型：期刊文献

英文题名：PToco: Prefix-based Token-level Collaboration Enhances Reasoning for Multi-LLMs

作者：Bian, Yuang[1]; Lin, Yupian[1]; Liu, Jingping[1]; Ruan, Tong[1]

机构：[1] School of Information Science and Engineering, East China University of Science and Technology, Shanghai, China

年份：2025

起止页码：8326

外文期刊名：Proceedings - International Conference on Computational Linguistics, COLING

收录：EI(收录号：20250917954564)

语种：英文

外文关键词：Multi agent systems - Natural language processing systems

摘要：Collaboration between multiple Large Language Models (LLMs) has attracted significant attention for its potential to mitigate hallucinations and enhance reasoning capabilities. Previous approaches, such as multi-agent debate and decoding-time integration, either rely on highly capable models with strong self-reflection abilities or are limited to models sharing the same tokenizer. To address these limitations, we introduce PToco (Prefix-based Token-level Collaboration), a novel mechanism that enables effective collaboration among less capable LLMs, independent of tokenizer differences. PToco uses a prefix-grouping method to extract consensus among tokens with varying levels of granularity, ensuring coherent and robust token generation across multiple models. Experimental results on a series of reasoning tasks demonstrate that PToco significantly improves performance over individual models. Furthermore, this approach generalizes well across different quantities and sizes of participating models, providing a more flexible and efficient solution for multi-LLM ensembles. ? 2025 Association for Computational Linguistics.