Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Short-form Question Answering on Natural Questions (NQ) (test)
Loading...
36
EM
VSPO + PRS_short
2.876
11.4755
20.075
28.6745
Dec 8, 2025
EM
Updated 4d ago
Evaluation Results
Method
Method
Links
EM
VSPO + PRS_short
Backbone=Qwen2.5-3B-In...
2025.12
36
GRPO + PRS_short
Backbone=Qwen2.5-3B-In...
2025.12
35.18
PPO + PRS_short
Backbone=Qwen2.5-3B-In...
2025.12
32.1
SFT
Backbone=Qwen2.5-3B-In...
2025.12
24.9
Untrained
Backbone=Qwen2.5-3B-In...
2025.12
4.15
Feedback
Search any
task
Search any
task