Understanding Reasoning Ability of Language Models From the Perspective of Reasoning Paths Aggregation
This work addresses the problem of understanding and enhancing reasoning in language models for AI researchers, though it is incremental in building on existing path-based methods.
The paper investigates how pre-trained language models perform reasoning by proposing that they aggregate indirect reasoning paths seen during pre-training, formalized as random walks on knowledge or reasoning graphs. Experiments on logic and chain-of-thought reasoning datasets show that augmenting training with random walk paths can improve multi-step reasoning performance.
Pre-trained language models (LMs) are able to perform complex reasoning without explicit fine-tuning. To understand how pre-training with a next-token prediction objective contributes to the emergence of such reasoning capability, we propose that we can view an LM as deriving new conclusions by aggregating indirect reasoning paths seen at pre-training time. We found this perspective effective in two important cases of reasoning: logic reasoning with knowledge graphs (KGs) and chain-of-thought (CoT) reasoning. More specifically, we formalize the reasoning paths as random walk paths on the knowledge/reasoning graphs. Analyses of learned LM distributions suggest that a weighted sum of relevant random walk path probabilities is a reasonable way to explain how LMs reason. Experiments and analysis on multiple KG and CoT datasets reveal the effect of training on random walk paths and suggest that augmenting unlabeled random walk reasoning paths can improve real-world multi-step reasoning performance. code: https://github.com/WANGXinyiLinda/LM_random_walk