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

ScienceQA

Benchmarks

Task NameDataset NameSOTA ResultTrend
Science Question AnsweringScienceQA
Accuracy99.8
791
Visual Question AnsweringScienceQA
Accuracy98.18
446
Science Question AnsweringScienceQA IMG
Accuracy98.2
335
Science Question AnsweringScienceQA (SQA)
Accuracy88.8
273
Science Question AnsweringScienceQA (test)
Average Accuracy98.3
273
Scientific Question AnsweringScienceQA image
Accuracy98
259
Science Question AnsweringScienceQA SQA-IMG
Accuracy92.8
186
Multimodal Science Question AnsweringScienceQA IMG
Accuracy95.78
152
Visual Question AnsweringScienceQA img
Accuracy86.2
135
Science Question AnsweringScienceQA SQA-I
Accuracy96.6
122
Visual Question AnsweringScienceQA (test)
Accuracy98.2
115
Question AnsweringScienceQA
Accuracy96.3
96
Science Question AnsweringScienceQA IMG (test)
Accuracy98.4
74
Multimodal Question AnsweringScienceQA (test)
Accuracy92.53
73
Science Question AnsweringScienceQA Image
Score95.6
70
Science Question AnsweringScienceQA
IMG Score88.6
64
Scientific Question AnsweringScienceQA
Accuracy88.6
61
Multimodal Question AnsweringScienceQA
Accuracy97.8
61
Visual Question AnsweringScienceQA image
Accuracy75.1
58
Scientific ReasoningScienceQA
Score96.09
54
Multimodal ReasoningScienceQA
Average Accuracy97.85
45
Visual Question AnsweringScienceQA Image (test)
Accuracy96.2
45
Visual Question AnsweringScienceQA (SQA)
SQA Accuracy89.3
43
Out-of-scope refusalScienceQA out-of-scope (test)
Refusal Rate0
40
Multimodal Science Question AnsweringScienceQA
Overall Average Score88.4
36
Showing 25 of 96 rows