Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Zero-shot Commonsense Reasoning on PIQA, HellaSwag, WinoGrande, ARC-Easy, OpenBookQA, and MathQA (test)
Loading...
59
Zero-shot Accuracy
Original
0.76
15.88
31
46.12
May 15, 2026
Zero-shot Accuracy
Updated 16d ago
Evaluation Results
Method
Method
Links
Zero-shot Accuracy
Original
Model Backbone=LLaMA-1...
2026.05
59
Original
Model Backbone=Vicuna-...
2026.05
56
ZS-SVD
Model Backbone=LLaMA-1...
2026.05
56
IO-SVD
Model Backbone=LLaMA-1...
2026.05
56
SVDLLM
Model Backbone=LLaMA-1...
2026.05
55
ZS-SVD
Model Backbone=Vicuna-...
2026.05
54
ASVD
Model Backbone=LLaMA-1...
2026.05
54
IO-SVD
Model Backbone=Vicuna-...
2026.05
53
Original
Model Backbone=OPT-6.7...
2026.05
52
ZS-SVD
Model Backbone=OPT-6.7...
2026.05
51
IO-SVD
Model Backbone=OPT-6.7...
2026.05
51
SVDLLM
Model Backbone=Vicuna-...
2026.05
51
FWSVD
Model Backbone=LLaMA-1...
2026.05
43
SVDLLM
Model Backbone=OPT-6.7...
2026.05
41
ASVD
Model Backbone=Vicuna-...
2026.05
33
ASVD
Model Backbone=OPT-6.7...
2026.05
32
SVD
Model Backbone=LLaMA-1...
2026.05
21
FWSVD
Model Backbone=Vicuna-...
2026.05
9
FWSVD
Model Backbone=OPT-6.7...
2026.05
6
SVD
Model Backbone=Vicuna-...
2026.05
5
SVD
Model Backbone=OPT-6.7...
2026.05
3
Feedback
Search any
task
Search any
task