Can Vision Language Models Judge Action Quality? An Empirical Evaluation

TL;DR

This paper evaluates the performance of state-of-the-art Vision Language Models in Action Quality Assessment across various domains, revealing significant limitations and biases that hinder reliable fine-grained movement evaluation.

Contribution

The study provides a comprehensive empirical evaluation of VLMs in AQA, highlighting their marginal performance and systematic biases, and establishes a baseline for future research.

Findings

VLMs perform only marginally above chance in AQA tasks.

Incorporating additional information yields isolated, inconsistent gains.

Models exhibit biases towards predicting correct execution regardless of visual evidence.

Abstract

Action Quality Assessment (AQA) has broad applications in physical therapy, sports coaching, and competitive judging. Although Vision Language Models (VLMs) hold considerable promise for AQA, their actual performance in this domain remains largely uncharacterised. We present a comprehensive evaluation of state-of-the-art VLMs across activity domains (e.g. fitness, figure skating, diving), tasks, representations, and prompting strategies. Baseline results reveal that Gemini 3.1 Pro, Qwen3-VL and InternVL3.5 models perform only marginally above random chance, and although strategies such as incorporation of skeleton information, grounding instructions, reasoning structures and in-context learning lead to isolated gains, none is consistently effective. Analysis of prediction distributions uncovers two systematic biases: a tendency to predict correct execution regardless of visual evidence, and a sensitivity to superficial linguistic framing. Reformulating tasks contrastively to mitigate these biases yields minimal improvement, suggesting that the models' limitations go beyond these biases, pointing to a fundamental difficulty with fine-grained movement quality assessment. Our findings establish a rigorous baseline for future VLM-based AQA research and provide an actionable outline for failure modes requiring mitigation prior to reliable real-world deployment.