Mirror: A Multiple-perspective Self-Reflection Method for Knowledge-rich Reasoning

TL;DR

Mirror introduces a multi-perspective self-reflection approach for LLMs to improve knowledge-rich reasoning by encouraging diverse reasoning paths and agreement among responses, outperforming existing methods.

Contribution

The paper proposes Mirror, a novel multi-perspective self-reflection method combining a Navigator and Reasoner to enhance reasoning diversity and reliability without ground truth access.

Findings

Outperforms several contemporary self-reflection methods on five reasoning datasets.

Strategies effectively address LLMs' struggles with revisiting predictions and self-assessment.

Ablation studies confirm the importance of diversity and agreement strategies.

Abstract

While Large language models (LLMs) have the capability to iteratively reflect on their own outputs, recent studies have observed their struggles with knowledge-rich problems without access to external resources. In addition to the inefficiency of LLMs in self-assessment, we also observe that LLMs struggle to revisit their predictions despite receiving explicit negative feedback. Therefore, We propose Mirror, a Multiple-perspective self-reflection method for knowledge-rich reasoning, to avoid getting stuck at a particular reflection iteration. Mirror enables LLMs to reflect from multiple-perspective clues, achieved through a heuristic interaction between a Navigator and a Reasoner. It guides agents toward diverse yet plausibly reliable reasoning trajectory without access to ground truth by encouraging (1) diversity of directions generated by Navigator and (2) agreement among strategically induced perturbations in responses generated by the Reasoner. The experiments on five reasoning datasets demonstrate that Mirror's superiority over several contemporary self-reflection approaches. Additionally, the ablation study studies clearly indicate that our strategies alleviate the aforementioned challenges.