Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

InfoQA

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multi-hop Question AnsweringInfoQA Synthetic Multi-hop QA 1-hop
Average F1100
24
Document Visual Question AnsweringInfoQA 105 (test)
Score86.9
23
Visual Question AnsweringInfoQA (test)
Accuracy83.1
19
Multi-hop Question AnsweringInfoQA Synthetic Multi-hop QA (2–4 hop)
Avg F198
12
Infographic Visual Question AnsweringInfoQA
ANLS73.43
11
Multi-hop Question AnsweringInfoQA Synthetic Multi-hop QA (4-hop)
Avg F180
4
Multi-hop Question AnsweringInfoQA Synthetic Multi-hop QA 3-hop
Avg F1 Score98
4
Showing 7 of 7 rows