Update README.md

README.md

@@ -35,6 +35,8 @@ Deeptheorem achieves the #Rank 5 position along all the commerical models and op
 | o1                     | 66\.67   | 61\.00  | 47\.46   | 47\.30  | 62\.50     | 57\.55  | 58\.88(\#3)   | 55\.28(\#2)  |
 | o3-mini                | 80\.95   | 77\.61  | 45\.76   | 43\.47  | 78\.12     | 75\.12  | 68\.28(\#1)   | 65\.40(\#1)  |
 | *DeepTheorem-RL-7B     | 55\.56   | 39\.07  | 28\.81   | 20\.85  | 57\.29     | 42\.20  | 47\.22(\#5)   | 34\.04(\#5)  |
+| *DeepTheorem-RL-3B     | 38\.10   | 23\.39  | 25\.42   | 13\.56  | 52\.08     | 33\.84  | 38\.53        | 23.60        | 
+| *DeepTheorem-RL-1.5B   | 31\.75   | 15\.23  | 23\.73   | 10\.15  | 52\.08     | 22\.79  | 35\.85        | 16.06        | 
 
 
 ## DeepTheorem Dataset 📊