Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Fact Retrieval on AcademicEval Abstract-multi
Loading...
27.5
F1 Score
Thought-Retriever
12.94
16.72
20.5
24.28
Apr 14, 2026
F1 Score
Win Rate
Updated 3d ago
Evaluation Results
Method
Method
Links
F1 Score
Win Rate
Thought-Retriever
retriever=Contriever,...
2026.04
27.5
50
Qwen3-Embed-8b
type=retriever-based,...
2026.04
24
20
IRCoT
type=retriever-based,...
2026.04
23.5
18
BM25
type=retriever-based,...
2026.04
23.2
7
Contriever
type=retriever-based,...
2026.04
23.2
15
DPR
type=retriever-based,...
2026.04
22.6
4
DRAGON
type=retriever-based,...
2026.04
22.6
8
TF-IDF
type=retriever-based,...
2026.04
22.5
4
Nous Hermes
context_window=32k
2026.04
20.4
7
RECOMP
type=retriever-based,...
2026.04
20.2
8
Full Context
variant=left
2026.04
15.5
0
Full Context
variant=right
2026.04
14.9
0
OpenOrca
context_window=8k
2026.04
13.5
3
Feedback
Search any
task
Search any
task