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

Medical Visual Question Answering research benchmarks – Page 5 · SOTA2 Research

Medical Visual Question Answering

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

92.8Plane Score

9

Feb 26, 2026

VQA-Med 2019 n=350 (test)

Gemini-3.5-Flash

68.3Accuracy (LLM Judge)

7

Jul 3, 2026

SLAKE n=350 (test)

86.3LLM Judge Accuracy

7

Jul 3, 2026

VQA-RAD n=350 (test)

Gemini-3.5-Flash

77.7LLM-Judge Accuracy

7

Jul 3, 2026

Medical-Diff-VQA 1.0 (test)

Attention-guided encoder-decoder

0.747BLEU-1

7

Jun 8, 2026

Trajectory-Aware Process Supervision

0.695BLEU-1

7

May 7, 2026

87.05Accuracy

7

Feb 26, 2026

VQA 2019 (test)

79.87Overall Accuracy

7

Feb 26, 2026

26.9Accuracy

6

Jul 1, 2026

59.7Accuracy

6

Mar 25, 2026

54.2Accuracy

6

Mar 25, 2026

94.4L-VASE

6

Mar 25, 2026

Modality-Heterogeneous Federated Medical VQA Task 2 (eight modality-specific clients)

90.02Client 1 Performance

6

Mar 24, 2026

VQA cross-domain 2019

77Accuracy

6

Feb 26, 2026

SLAKE cross-domain

74.6Accuracy

6

Feb 26, 2026

Medical Multi-Modal Task Suite

73Overall Score

6

Feb 26, 2026

Medical Multi-modal Task Aggregate (test)

0.729Overall Score

6

Feb 26, 2026

MIMOSeg Perspective IV

0.397ROUGE-L

6

Feb 26, 2026

MIMOSeg Perspective III

0.65ROUGE-L

6

Feb 26, 2026

MIMOSeg Perspective II

0.436ROUGE-L

6

Feb 26, 2026

PathVQA (held-out)

HuatuoGPT-Vision

59.9Accuracy

6

Feb 26, 2026

SLAKE (held-out)

HuatuoGPT-Vision

74.5Accuracy

6

Feb 26, 2026

VQA-RAD (held-out)

HuatuoGPT-Vision

63.8Accuracy

6

Feb 26, 2026

OmniMedVQA (ID)

5ECE

5

Jun 26, 2026

PMC-VQA ID (test)

0.07ECE

5

Jun 26, 2026

Page 5 of 8

Popular datasets

SLAKE VQA-RAD PMC-VQA VQA-Med PathVQA

Follow for updates

Join SOTA2 Contact