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

Multimodal Abstractive Summarization research benchmarks · SOTA2 Research

Multimodal Abstractive Summarization

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

MM-Sum mid-high-resource

41.51ROUGE-1

90

Feb 27, 2026

MM-Sum mid-high-resource (test)

41.59ROUGE-1

90

Feb 27, 2026

67.7ROUGE-1

21

May 13, 2026

MM-Sum Avg., low-resource

30.55ROUGE-1 Score

6

Feb 27, 2026

MM-Sum Welsh low-resource

32.86ROUGE-1

6

Feb 27, 2026

MM-Sum Telugu, low-resource

19.87ROUGE-1 Score

6

Feb 27, 2026

MM-Sum Swahili low-resource

38.02ROUGE-1

6

Feb 27, 2026

MM-Sum Serbian Latin, low-resource

21.89ROUGE-1

6

Feb 27, 2026

MM-Sum Serbian Cyrillic, low-resource

23.88R-1

6

Feb 27, 2026

MM-Sum Punjabi, low-resource

30.57ROUGE-1

6

Feb 27, 2026

MM-Sum Pidgin low-resource

37.91R-1

6

Feb 27, 2026

MM-Sum Pashto low-resource

38.15ROUGE-1

6

Feb 27, 2026

MM-Sum Oromo low-resource

18.82ROUGE-1

6

Feb 27, 2026

MM-Sum Marathi low-resource

22.61ROUGE-1

6

Feb 27, 2026

MM-Sum Japanese low-resource

47.79ROUGE-1 Score

6

Feb 27, 2026

MM-Sum Hausa low-resource

39.28ROUGE-1

6

Feb 27, 2026

MM-Sum Gujarati, low-resource

22.18ROUGE-1

6

Feb 27, 2026

MM-Sum French, low-resource

35.93ROUGE-1

6

Feb 27, 2026

MM-Sum Bengali low-resource

28.58ROUGE-1

6

Feb 27, 2026

Chinese human evaluation set

3.98Fluency

3

Feb 27, 2026

English human evaluation set

4.56Fluency

3

Feb 27, 2026

MM-Sum mid-high-resource 1.0

—

—ROUGE-1

0

Feb 26, 2026

Popular datasets

MM-Sum How2 English human evaluation Chinese human evaluation set

Follow for updates

Join SOTA2 Contact