Planning on Trips (test)
15.6AccuracyReProbe
Evaluation Results
| Method | Links | |
|---|---|---|
| ReProbe# Sample=32K, Input Features=Hidden States, Annotation Source=Self-anno, Decoding Method=Beam search2025.11 | 15.6 | |
| ReProbe# Sample=32K, Input Features=Attn+Logit, Annotation Source=Self-anno, Decoding Method=Beam search2025.11 | 15 | |
| ReProbe# Sample=32K, Input Features=Hidden States, Annotation Source=DeepSeek-anno, Decoding Method=Beam search2025.11 | 13.4 | |
| Qwen2.5-Math-7B-PRM800k# Sample=263K, Decoding Method=Beam search2025.11 | 13.1 | |
| Qwen2.5-Math-PRM-7B# Sample=860K, Decoding Method=Beam search2025.11 | 13.1 | |
| ReProbe# Sample=32K, Input Features=Attn+Logit, Annotation Source=DeepSeek-anno, Decoding Method=Beam search2025.11 | 11.8 |