Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Abstention on AbstentionBench BB Known Unk.
Loading...
95.8
Abstention F1
TIAR
83.944
87.022
90.1
93.178
May 25, 2026
Abstention F1
Abstention Recall
Abstention Precision
Accuracy
Updated 8d ago
Evaluation Results
Method
Method
Links
Abstention F1
Abstention Recall
Abstention Precision
Accuracy
TIAR
Backbone=Llama-3.1-8B-...
2026.05
95.8
100
92
82.6
Claude Sonnet 4.5
Model Type=Proprietary...
2026.05
93.9
100
88.5
82.6
GPT-5.2
Model Type=Proprietary...
2026.05
93.6
95.7
91.7
82.6
Gemini 3
Model Type=Proprietary...
2026.05
84.4
82.6
86.4
82.6
Feedback
Search any
task
Search any
task