updated spell backward impl

reasoning_gym/algorithmic/spell_backward.py

@@ -3,7 +3,7 @@
 import re
 from dataclasses import dataclass
 from random import Random
-from typing import Optional
+from typing import Any, Dict, Optional
 
 from ..data import read_data_file
 from ..factory import ProceduralDataset, register_dataset
@@ -49,5 +49,18 @@ class SpellBackwardDataset(ProceduralDataset):
             "metadata": {"word": word, "word_len": len(word)},
         }
 
+    def score_answer(self, answer: Optional[str], entry: Dict[str, Any]) -> float:
+        reward = 0
+        expected_answer = entry["answer"]
+        if answer is not None:
+            try:
+                if expected_answer.lower() == answer.lower():
+                    reward = 1.0
+                else:
+                    reward = 0.05
+            except:
+                reward = 0.01
+        return reward
+
 
 register_dataset("spell_backward", SpellBackwardDataset, SpellBackwardConfig)