Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Video Question Answering research benchmarks – Page 16 · SOTA2 Research

Video Question Answering

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

NEXT-QA open-form generation

34.3WUPS

5

Feb 26, 2026

Causal-VidQA 1.0 (test)

74.36Acc@P

5

Feb 26, 2026

AGQA (more compositional steps)

52.24Binary Accuracy

5

Feb 26, 2026

AGQA (novel compositions)

49.27Binary Accuracy

5

Feb 26, 2026

52.7Accuracy

5

Feb 26, 2026

70.9Accuracy

5

Feb 26, 2026

63.5Accuracy

5

Feb 26, 2026

TemporalBench MBA-short QA

36.7Multi-binary Acc

5

Feb 26, 2026

44.5Paired Accuracy

5

Feb 26, 2026

69.25AccV

4

Jun 30, 2026

45.45Accuracy V

4

Jun 30, 2026

Video-MME short and medium videos

50.1Temporal Accuracy

4

Jun 23, 2026

69.7Accuracy

4

Jun 23, 2026

55.1LLM Judge Score

4

Jun 4, 2026

0.493LLM Judge Score

4

Jun 4, 2026

62.7LLM Judge Score

4

Jun 4, 2026

61LLM Judge Score

4

Jun 4, 2026

MovieChat (test)

CogniGPT (Ours)

95.3Accuracy

4

May 7, 2026

Video-MME (VMME) 10-hour variant (no subtitles) 1.0 (test)

69.1Accuracy

4

Mar 19, 2026

LongVideoBench (LVB) 10-hour variant 1.0 (test)

70.1Accuracy

4

Mar 19, 2026

How2QA public (val)

84.4Top-1 Acc

4

Feb 27, 2026

VideoEspresso (test)

REVISE (Reinforced Fine-Tuning)

0.278Accuracy

4

Feb 26, 2026

EgoSchema ES-S (public subset)

55.8Accuracy

4

Feb 26, 2026

IntentQA original (test)

video-grounded entailment tree reasoning

66.1Temporal Accuracy

4

Feb 26, 2026

NExT-QA original (test)

67Temporal Score

4

Feb 26, 2026

Page 16 of 18

Popular datasets

Video-MME VideoMME NExT-QA EgoSchema TGIF-QA

Follow for updates

Join SOTA2 Contact