Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Formulation QA on Formulation QA (OOD)
Loading...
39.4
Accuracy
Original
33.0872
34.7261
36.365
38.0039
Apr 8, 2026
Accuracy
Updated 9d ago
Evaluation Results
Method
Method
Links
Accuracy
Original
Token=0, Backbone=LLaM...
2026.04
39.4
w GPT-5
Token=50*3, Backbone=L...
2026.04
38.3
SciDC
Token=30*5, Backbone=L...
2026.04
38.3
w wiki
Token=800*5, Backbone=...
2026.04
36.1
vanilla RAG
Token=512*3, Backbone=...
2026.04
35.6
w summary
Token=30*5, Backbone=L...
2026.04
33.33
Feedback
Search any
task
Search any
task