Benchmark RadarAI BENCHMARK PROFILE
FitAQA
FitAQA evaluates fitness action quality assessment in MLLMs across perception, judgement, and temporal grounding tasks, using a unified taxonomy of 38 form errors in six dimensions over 30 exercises.
- Released
- 2026-08-09
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
It provides a systematic benchmark for fitness AQA, enabling assessment of MLLMs' ability to perceive and reason about exercise quality, and identifies visual perception as a key bottleneck.
Motivation
Fitness Action Quality Assessment (AQA) is important for intelligent sports training, yet the capabilities of Multimodal Large Language Models (MLLMs) in this setting remain underexplored.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.