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

Short-form Question Answering research benchmarks · SOTA2 Research

Short-form Question Answering

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

TQA (TriviaQA) (test)

68.4F1 Score

18

Jun 12, 2026

7.9ECE

18

Mar 4, 2026

40F1 Score

14

Jul 3, 2026

SAGE Web Search

63.3Accuracy (Com. Sci.)

12

Feb 26, 2026

CO-LMLM-360M-FW

21.7Accuracy

11

Jul 9, 2026

CO-LMLM-360M-FW

50.6Accuracy

11

Jul 9, 2026

HF/SMOLLM2-1.7B*

45.1Accuracy

11

Jul 9, 2026

77.3F1 Score

9

Jun 12, 2026

Rewarding Doubt*

85.9AUROC

7

May 15, 2026

Short-form QA Aggregate (Avg.) (test)

VSPO + PRS_short

35.93EM

5

Feb 26, 2026

Bamboogle (test)

VSPO + PRS_short

0.349EM

5

Feb 26, 2026

VSPO + PRS_short

13.67EM

5

Feb 26, 2026

PPO + PRS_short

26.5EM

5

Feb 26, 2026

VSPO + PRS_short

36.34Exact Match

5

Feb 26, 2026

TriviaQA (test)

VSPO + PRS_short

50Exact Match

5

Feb 26, 2026

Natural Questions (NQ) (test)

VSPO + PRS_short

36EM

5

Feb 26, 2026

Popular datasets

TriviaQA SimpleQA PopQA Trivia QA MuSiQue

Follow for updates

Join SOTA2 Contact