MD-VQA Benchmark

What is MD-VQA?

Evaluates mistake detection in videos via question answering, testing whether methods can discern if a step was executed correctly against its description for…

Released

2026-08-28

Evaluates

General AI, Vision & 3D, cs.CV

Openness

unknown

Importer

Claire Radar

Review status

ai-name-audit-deferred

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2608.28406

MD-VQA paper and code

• Paper https://arxiv.org/abs/2608.28406
• Code https://github.com/FedeSpu/mstk

Which sources cite MD-VQA?

• MD-VQA

Related General AI, Vision & 3D, cs.CV benchmarks

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• 360CityArena
• 3DCodeBench
• 5-Dialects-BN
• A3S-Bench
• AARRI-Bench
• ActBench
• ActiveSciBench