Don't "Overthink" Passage Reranking: Is Reasoning Truly Necessary?

TL;DR

This study questions the necessity of reasoning in passage reranking with LLMs, showing that non-reasoning methods often outperform reasoning-based approaches due to limitations in the reasoning process.

Contribution

The paper provides a comparative analysis demonstrating that reasoning-based rerankers do not necessarily improve accuracy and can be less effective than standard rerankers, challenging current assumptions.

Findings

Standard rerankers outperform reasoning-based rerankers.

Disabling reasoning improves reranking effectiveness.

Reasoning pushes relevance scores toward polarization.

Abstract

With the growing success of reasoning models across complex natural language tasks, researchers in the Information Retrieval (IR) community have begun exploring how similar reasoning capabilities can be integrated into passage rerankers built on Large Language Models (LLMs). These methods typically employ an LLM to produce an explicit, step-by-step reasoning process before arriving at a final relevance prediction. But, does reasoning actually improve reranking accuracy? In this paper, we dive deeper into this question, studying the impact of the reasoning process by comparing reasoning-based pointwise rerankers (ReasonRR) to standard, non-reasoning pointwise rerankers (StandardRR) under identical training conditions, and observe that StandardRR generally outperforms ReasonRR. Building on this observation, we then study the importance of reasoning to ReasonRR by disabling its reasoning process (ReasonRR-NoReason), and find that ReasonRR-NoReason is surprisingly more effective than ReasonRR. Examining the cause of this result, our findings reveal that reasoning-based rerankers are limited by the LLM's reasoning process, which pushes it toward polarized relevance scores and thus fails to consider the partial relevance of passages, a key factor for the accuracy of pointwise rerankers.