Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multi-Hop Search-augmented Question Answering on 2Wiki (success rate %)
Loading...
35.6
Success Rate
+ MASA
22.288
25.744
29.2
32.656
May 29, 2026
Success Rate
Updated 2d ago
Evaluation Results
Method
Method
Links
Success Rate
+ MASA
Backbone=Qwen3-32B
2026.05
35.6
+ DS-Adapter
Backbone=Qwen3-32B
2026.05
32
No Skill
Backbone=Qwen3-8B
2026.05
30.6
+ Base Skill
Backbone=Qwen3-14B
2026.05
30.3
+ MASA
Backbone=Qwen3-14B
2026.05
30
No Skill
Backbone=Qwen3-32B
2026.05
29.3
+ DS-Adapter
Backbone=Qwen3-14B
2026.05
28.5
+ MASA
Backbone=Qwen3-4B
2026.05
27
No Skill
Backbone=Qwen3-14B
2026.05
26.8
+ Base Skill
Backbone=Qwen3-32B
2026.05
26
+ Base Skill
Backbone=Qwen3-8B
2026.05
25.9
+ MASA
Backbone=Qwen3-8B
2026.05
25.7
+ Base Skill
Backbone=Qwen3-4B
2026.05
24.4
+ DS-Adapter
Backbone=Qwen3-4B
2026.05
23.9
+ DS-Adapter
Backbone=Qwen3-8B
2026.05
22.9
No Skill
Backbone=Qwen3-4B
2026.05
22.8
Feedback
Search any
task
Search any
task