Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Fact Verification on FEVER (EM)
Loading...
61.1
EM
SAVER
48.828
52.014
55.2
58.386
Apr 9, 2026
EM
Updated 8d ago
Evaluation Results
Method
Method
Links
EM
SAVER
Backbone=LLaMA-3.1-8B,...
2026.04
61.1
SAVER
Backbone=Qwen-2.5-7B,...
2026.04
60.9
MAD
Backbone=LLaMA-3.1-8B,...
2026.04
60.7
B-2
Backbone=Qwen-2.5-7B,...
2026.04
60.7
B-2
Backbone=LLaMA-3.1-8B,...
2026.04
60.6
MAD
Backbone=Qwen-2.5-7B,...
2026.04
60.1
Self-Refine
Backbone=Qwen-2.5-7B,...
2026.04
58.2
Self-Refine
Backbone=LLaMA-3.1-8B,...
2026.04
57.6
SAVER
Backbone=LLaMA-3.2-3B,...
2026.04
56.4
CoT
Backbone=Qwen-2.5-7B,...
2026.04
56.3
CoT
Backbone=LLaMA-3.1-8B,...
2026.04
55.9
MAD
Backbone=LLaMA-3.2-3B,...
2026.04
55.6
B-2
Backbone=LLaMA-3.2-3B,...
2026.04
54.3
Self-Refine
Backbone=LLaMA-3.2-3B,...
2026.04
53.8
Vanilla LM
Backbone=LLaMA-3.1-8B,...
2026.04
53.7
Vanilla LM
Backbone=Qwen-2.5-7B,...
2026.04
52.7
CoT
Backbone=LLaMA-3.2-3B,...
2026.04
52.4
Vanilla LM
Backbone=LLaMA-3.2-3B,...
2026.04
49.3
Feedback
Search any
task
Search any
task