PRIME-RL
/

EurusPRM-Stage2

Model card Files Files and versions

yuchenFan commited on Jan 16

Commit

46d8117

·

verified ·

1 Parent(s): e078587

Update README.md

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -186,7 +186,7 @@ The threshold is obtained by converting the original score of each step using si
 For leveraging the capibility of **EurusPRM** better, we add ``Step K`` (where K is the actual index of the step) in front of each step in **ProcessBench**.
 | Reward Model | GSM8k | MATH | OlympiadBench | Omni-Math | Avg |
-| --- | --- | --- | --- | --- | --- | --- | --- |
 | Math-Shepherd-PRM-7B | 47.9 | 29.5 | 24.8 | 23.8 | 31.5 |
 | RLHFlow-PRM-Mistral-8B | 50.4 | 33.4 | 13.8 | 15.8 | 28.4 |
 | RLHFlow-PRM-Deepseek-8B | 38.8 | 33.8 | 16.9 | 16.9 | 26.6 |

 For leveraging the capibility of **EurusPRM** better, we add ``Step K`` (where K is the actual index of the step) in front of each step in **ProcessBench**.
 | Reward Model | GSM8k | MATH | OlympiadBench | Omni-Math | Avg |
+| --- | --- | --- | --- | --- | --- | --- |
 | Math-Shepherd-PRM-7B | 47.9 | 29.5 | 24.8 | 23.8 | 31.5 |
 | RLHFlow-PRM-Mistral-8B | 50.4 | 33.4 | 13.8 | 15.8 | 28.4 |
 | RLHFlow-PRM-Deepseek-8B | 38.8 | 33.8 | 16.9 | 16.9 | 26.6 |