Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Chain-of-Thought Faithfulness Evaluation on Math Book
Loading...
22.8
Unfaithful Explanation Rate
DPO + RMC
22.304
25.652
29
32.348
Apr 7, 2025
Unfaithful Explanation Rate
Updated 18d ago
Evaluation Results
Method
Method
Links
Unfaithful Explanation Rate
DPO + RMC
Reward Model=SK-LLAMA-...
2025.04
22.8
DPO + RMC
Reward Model=SK-GEMMA-...
2025.04
23.6
DPO + RMD
Reward Model=SK-LLAMA-...
2025.04
24.5
Base LLAMA-3.1-8B-IT
Reward Model=None, Dec...
2025.04
24.8
DPO + RM
Reward Model=SK-LLAMA-...
2025.04
25.7
Base LLAMA-3.1-8B-IT
Reward Model=None, Dec...
2025.04
27.2
DPO + RM
Reward Model=SK-GEMMA-...
2025.04
27.2
DPO + RMD
Reward Model=SK-GEMMA-...
2025.04
28.3
DPO + RMC
Reward Model=SK-LLAMA-...
2025.04
31.6
DPO + RMC
Reward Model=SK-GEMMA-...
2025.04
32.5
DPO + RMD
Reward Model=SK-LLAMA-...
2025.04
33.6
DPO + RM
Reward Model=SK-GEMMA-...
2025.04
33.9
DPO + RM
Reward Model=SK-LLAMA-...
2025.04
34
DPO + RMD
Reward Model=SK-GEMMA-...
2025.04
35.2
Feedback
Search any
task
Search any
task