fix: aggregate Global Leaderboard by model to prevent duplicates

app.py

@@ -56,15 +56,21 @@ class LeaderboardManager:
         return self.leaderboard.copy()
     
     def get_top_results(self, n: int = None) -> pd.DataFrame:
-        """Get top N results by composite score."""
+        """Get top N results by composite score, aggregated by model."""
         if self.leaderboard.empty:
             return self.leaderboard
         
         if n is None:
             n = self.config.top_results
         
+        # Group by model and calculate averages
+        numeric_columns = ['composite_score', 'correctness_exact', 'result_match_f1', 'exec_success', 'latency_ms']
+        
+        # Calculate averages for numeric columns
+        model_aggregated = self.leaderboard.groupby('model_name')[numeric_columns].mean().reset_index()
+        
         # Sort by composite score and add ranking
-        top_results = (self.leaderboard
+        top_results = (model_aggregated
                       .sort_values('composite_score', ascending=False)
                       .head(n)
                       .reset_index(drop=True))