VersusQ: Pairwise Margin Reasoning for Generalizable Video Quality Assessment

TL;DR

VersusQ introduces a pairwise comparison framework for video quality assessment using large multimodal models, improving cross-domain generalization and ranking accuracy by focusing on perceptual differences rather than absolute scores.

Contribution

It proposes VersusQ, a novel pairwise margin reasoning method that enhances generalization and ranking in video quality assessment by leveraging direct comparisons.

Findings

Achieves state-of-the-art performance on multiple benchmarks.

Demonstrates strong cross-domain generalization.

Provides reliable fine-grained ranking in diverse scenarios.

Abstract

Large Multimodal Models (LMMs) have shown promise for video quality assessment, but most methods still predict an absolute score for each video. Such pointwise supervision often mixes perceptual quality with dataset-specific calibration, including annotation protocols, rating habits, and score distributions. As a result, the learned scoring rule may work well within a benchmark but transfer poorly across unseen domains. We argue that relative comparisons alleviate the absolute-scale calibration bias by focusing purely on perceptual differences rather than dataset-specific rating habits. Consequently, we propose \textbf{VersusQ}, a pairwise margin reasoning framework driven entirely by direct comparisons. Specifically, VersusQ performs LMM-based comparison between two videos, reasons about their visual and temporal quality differences, and predicts a signed continuous margin that captures both the preferred choice and the degree of difference. Furthermore, to align interpretable comparison rationales with fine-grained numerical differences, we introduce Margin-Coupled GRPO, which jointly optimizes rollout-based relational reasoning and continuous margin regression. Extensive experiments on multiple public VQA benchmarks demonstrate that VersusQ achieves state-of-the-art performance, strong cross-domain generalization, and reliable fine-grained ranking under heterogeneous evaluation scenarios.