Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multi-Hop Search-augmented Question Answering on MuSiQue (success rate %)
Loading...
11.8
Success Rate
+ MASA
5.352
7.026
8.7
10.374
May 29, 2026
Success Rate
Updated 2d ago
Evaluation Results
Method
Method
Links
Success Rate
+ MASA
Backbone=Qwen3-32B
2026.05
11.8
+ Base Skill
Backbone=Qwen3-32B
2026.05
11.7
+ DS-Adapter
Backbone=Qwen3-32B
2026.05
11.6
+ Base Skill
Backbone=Qwen3-14B
2026.05
11.4
+ MASA
Backbone=Qwen3-8B
2026.05
10
+ MASA
Backbone=Qwen3-14B
2026.05
9.7
+ MASA
Backbone=Qwen3-4B
2026.05
9.4
+ DS-Adapter
Backbone=Qwen3-4B
2026.05
9.3
+ DS-Adapter
Backbone=Qwen3-14B
2026.05
9.2
No Skill
Backbone=Qwen3-32B
2026.05
8.6
+ Base Skill
Backbone=Qwen3-4B
2026.05
7.8
No Skill
Backbone=Qwen3-14B
2026.05
7.6
No Skill
Backbone=Qwen3-8B
2026.05
6.7
No Skill
Backbone=Qwen3-4B
2026.05
6.4
+ Base Skill
Backbone=Qwen3-8B
2026.05
6.2
+ DS-Adapter
Backbone=Qwen3-8B
2026.05
5.6
Feedback
Search any
task
Search any
task