Large Language Models Are Cross-Lingual Knowledge-Free Reasoners
This addresses the problem of understanding cross-lingual reasoning transfer in LLMs for NLP researchers, though it is incremental in analyzing existing capabilities.
The study decomposed reasoning into knowledge retrieval and knowledge-free components, finding that knowledge-free reasoning transfers nearly perfectly across languages, while cross-lingual knowledge retrieval hinders transfer, with analysis showing higher similarity in hidden states and neuron activation explains this difference.
Large Language Models have demonstrated impressive reasoning capabilities across multiple languages. However, the relationship between capabilities in different languages is less explored. In this work, we decompose the process of reasoning tasks into two separated components: knowledge retrieval and knowledge-free reasoning, and analyze the relationship between cross-lingual transferability and these two components. With adapted commonsense reasoning datasets and constructed knowledge-free reasoning datasets, we show that the knowledge-free reasoning capability can be nearly perfectly transferred across various source-target language directions despite the secondary impact of resource in some specific target languages, while cross-lingual knowledge retrieval significantly hinders the transfer. Moreover, by analyzing the hidden states and feed-forward network neuron activation during the reasoning, we show that higher similarity of hidden representations and larger overlap of activated neurons could explain the better cross-lingual transferability of knowledge-free reasoning than knowledge retrieval. Thus, we hypothesize that knowledge-free reasoning shares similar neurons in different languages for reasoning, while knowledge is stored separately in different languages. Our code and data is available at: https://github.com/NJUNLP/Knowledge-Free-Reasoning.