A Systematic Analysis of Large Language Models as Soft Reasoners: The Case of Syllogistic Inferences

TL;DR

This paper systematically investigates how large language models perform syllogistic reasoning, analyzing the effects of chain-of-thought prompting, in-context learning, and fine-tuning on reasoning accuracy and biases.

Contribution

It provides a comprehensive analysis of LLM reasoning in syllogisms, highlighting how different training methods affect reasoning biases and model consistency.

Findings

ICL and SFT improve reasoning accuracy

SFT reduces reasoning biases more effectively

Models exhibit heuristics similar to cognitive science findings

Abstract

The reasoning abilities of Large Language Models (LLMs) are becoming a central focus of study in NLP. In this paper, we consider the case of syllogistic reasoning, an area of deductive reasoning studied extensively in logic and cognitive psychology. Previous research has shown that pre-trained LLMs exhibit reasoning biases, such as $\textit{content effects}$, avoid answering that $\textit{no conclusion follows}$, display human-like difficulties, and struggle with multi-step reasoning. We contribute to this research line by systematically investigating the effects of chain-of-thought reasoning, in-context learning (ICL), and supervised fine-tuning (SFT) on syllogistic reasoning, considering syllogisms with conclusions that support or violate world knowledge, as well as ones with multiple premises. Crucially, we go beyond the standard focus on accuracy, with an in-depth analysis of the conclusions generated by the models. Our results suggest that the behavior of pre-trained LLMs can be explained by heuristics studied in cognitive science and that both ICL and SFT improve model performance on valid inferences, although only the latter mitigates most reasoning biases without harming model consistency.