vix.ing · top · new · best · stats

Faithful Chain-of-Thought Reasoning

2023/01/31 by Qing Lyu, Lyu, Qing, Shreya Havaldar +13 · 1 voice · 96 citations
Computer Science · #Advanced Graph Neural Networks #Natural Language Processing Techniques #Topic Modeling #cs.CL

paper · pdf · doi:10.48550/arxiv.2301.13379

arxiv published 2023/01/31 · arxiv updated 2023/09/20

Abstract

While Chain-of-Thought (CoT) prompting boosts Language Models' (LM) performance on a gamut of complex reasoning tasks, the generated reasoning chain does not necessarily reflect how the model arrives at the answer (aka. faithfulness). We propose Faithful CoT, a reasoning framework involving two stages: Translation (Natural Language query → symbolic reasoning chain) and Problem Solving (reasoning chain → answer), using an LM and a deterministic solver respectively. This guarantees that the reasoning chain provides a faithful explanation of the final answer. Aside from interpretability, Faithful CoT also improves empirical performance: it outperforms standard CoT on 9 of 10 benchmarks from 4 diverse domains, with a relative accuracy gain of 6.3% on Math Word Problems (MWP), 3.4% on Planning, 5.5% on Multi-hop Question Answering (QA), and 21.4% on Relational Inference. Furthermore, with GPT-4 and Codex, it sets the new state-of-the-art few-shot performance on 7 datasets (with 95.0+ accuracy on 6 of them), showing a strong synergy between faithfulness and accuracy.

Cited by

Discussions

Related