Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multi-step interaction on 20Q
Loading...
32.1
Winrate
NLAC
11.612
16.931
22.25
27.569
Dec 4, 2025
Winrate
Updated 1mo ago
Evaluation Results
Method
Method
Links
Winrate
NLAC
Paradigm=Fine-tuning,...
2025.12
32.1
NLRL
Paradigm=Fine-tuning,...
2025.12
31.8
NLQL
Paradigm=Fine-tuning,...
2025.12
31.4
ReAct
Paradigm=Prompting, Ba...
2025.12
30.2
Self-Distillation
Paradigm=Fine-tuning,...
2025.12
26.8
NLAC
Paradigm=Fine-tuning,...
2025.12
26
NLRL
Paradigm=Fine-tuning,...
2025.12
25.8
GRPO
Paradigm=Fine-tuning,...
2025.12
25.6
NLQL
Paradigm=Fine-tuning,...
2025.12
24.2
PPO
Paradigm=Fine-tuning,...
2025.12
24
RFT
Paradigm=Fine-tuning,...
2025.12
22
GRPO
Paradigm=Fine-tuning,...
2025.12
18.4
PPO
Paradigm=Fine-tuning,...
2025.12
17.2
RFT
Paradigm=Fine-tuning,...
2025.12
12.6
Self-Distillation
Paradigm=Fine-tuning,...
2025.12
12.4
Feedback
Search any
task
Search any
task
