MR. Judge: Multimodal Reasoner as a Judge

TL;DR

MR. Judge introduces a multimodal reasoning-based judging paradigm for large language models, enhancing interpretability and performance in response evaluation through deliberate reasoning and automatic annotation strategies.

Contribution

The paper proposes MR. Judge, a novel multimodal reasoning framework for LLM judges, with automatic annotation techniques to improve evaluation accuracy and interpretability.

Findings

MR. Judge-7B surpasses GPT-4o by 9.9% on VL-RewardBench.

Improves MM-Vet performance during inference-time scaling by up to 7.7%.

Effective across various tasks with enhanced reasoning capabilities.

Abstract

The paradigm of using Large Language Models (LLMs) and Multimodal Large Language Models (MLLMs) as evaluative judges has emerged as an effective approach in RLHF and inference-time scaling. In this work, we propose Multimodal Reasoner as a Judge (MR. Judge), a paradigm for empowering general-purpose MLLMs judges with strong reasoning capabilities. Instead of directly assigning scores for each response, we formulate the judgement process as a reasoning-inspired multiple-choice problem. Specifically, the judge model first conducts deliberate reasoning covering different aspects of the responses and eventually selects the best response from them. This reasoning process not only improves the interpretibility of the judgement, but also greatly enhances the performance of MLLM judges. To cope with the lack of questions with scored responses, we propose the following strategy to achieve automatic annotation: 1) Reverse Response Candidates Synthesis: starting from a supervised fine-tuning (SFT) dataset, we treat the original response as the best candidate and prompt the MLLM to generate plausible but flawed negative candidates. 2) Text-based reasoning extraction: we carefully design a data synthesis pipeline for distilling the reasoning capability from a text-based reasoning model, which is adopted to enable the MLLM judges to regain complex reasoning ability via warm up supervised fine-tuning. Experiments demonstrate that our MR. Judge is effective across a wide range of tasks. Specifically, our MR. Judge-7B surpasses GPT-4o by 9.9% on VL-RewardBench, and improves performance on MM-Vet during inference-time scaling by up to 7.7%.