Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Fact-checking on AdvFact
Loading...
64
Accuracy
ToE
5.76
20.88
36
51.12
Jun 26, 2026
Accuracy
Updated 26d ago
Evaluation Results
Method
Method
Links
Accuracy
ToE
Backbone=DeepSeek-V3.2
2026.06
64
ToE
Backbone=GPT-OSS-20B
2026.06
54
Adsent
Backbone=GPT-OSS-20B
2026.06
52
TELLER
Backbone=DeepSeek-V3.2
2026.06
40
ZCoT
Backbone=DeepSeek-V3.2
2026.06
36
TELLER
Backbone=GPT-OSS-20B
2026.06
32
Adsent
Backbone=DeepSeek-V3.2
2026.06
30
DefGen
Backbone=DeepSeek-V3.2
2026.06
28
DefGen
Backbone=GPT-OSS-20B
2026.06
24
ZCoT
Backbone=GPT-OSS-20B
2026.06
20
Direct
Backbone=DeepSeek-V3.2
2026.06
16
STEEL
Backbone=DeepSeek-V3.2
2026.06
16
STEEL
Backbone=GPT-OSS-20B
2026.06
14
AFacTA
Backbone=DeepSeek-V3.2
2026.06
12
AFacTA
Backbone=GPT-OSS-20B
2026.06
12
Direct
Backbone=GPT-OSS-20B
2026.06
8
Feedback
Search any
task
Search any
task