Faithful Chain-of-Thought Reasoning
This addresses the interpretability and accuracy issues in reasoning tasks for AI researchers and practitioners, though it is incremental as it builds on existing CoT methods.
The paper tackles the problem of unfaithful reasoning in Chain-of-Thought prompting for language models by proposing Faithful CoT, a two-stage framework that guarantees the reasoning chain explains the final answer, resulting in improved accuracy across multiple benchmarks, such as a 6.3% gain on Math Word Problems and 21.4% on Relational Inference.
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 $\rightarrow$ symbolic reasoning chain) and Problem Solving (reasoning chain $\rightarrow$ 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.