A reward model that multiplies stepwise correctness and potential scores improves best-of-N verification accuracy for math reasoning.
Reason from Fallacy: Enhancing Large Language Models' Logical Reasoning through Logical Fallacy Understanding
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
Large Language Models (LLMs) have demonstrated good performance in many reasoning tasks, but they still struggle with some complicated reasoning tasks including logical reasoning. One non-negligible reason for LLMs' suboptimal performance on logical reasoning is their overlooking of understanding logical fallacies correctly. To evaluate LLMs' capability of logical fallacy understanding (LFU), we propose five concrete tasks from three cognitive dimensions of WHAT, WHY, and HOW in this paper. Towards these LFU tasks, we have successfully constructed a new dataset LFUD based on GPT-4 accompanied by a little human effort. Our extensive experiments justify that our LFUD can be used not only to evaluate LLMs' LFU capability, but also to fine-tune LLMs to obtain significantly enhanced performance on logical reasoning.
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
DuaShepherd: Integrating Stepwise Correctness and Potential Rewards for Mathematical Reasoning
A reward model that multiplies stepwise correctness and potential scores improves best-of-N verification accuracy for math reasoning.