Merge remote-tracking branch 'github/main'

# Conflicts:
# README.md
# backend/evaluation/ablation_chunk_size.py
# backend/evaluation/backbone_comparison.py
# backend/evaluation/baseline_runner.py
# backend/evaluation/benchmark_runner.py
# backend/evaluation/metrics.py

README.md

@@ -7,7 +7,6 @@ sdk: docker
 pinned: false
 license: mit
 ---
-
 # MEXAR Ultimate 🧠
 
 **Multimodal Explainable AI Reasoning Assistant**

backend/evaluation/baseline_runner.py

@@ -4,13 +4,11 @@ Runs CRAG and RAPTOR baselines against a set of test queries.
 import sys
 import os
 from typing import Dict, List, Optional
-
 sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 from modules.reasoning_engine import create_reasoning_engine
 from evaluation.metrics import MetricsRunner
 
-
 def _append_score(results: Dict[str, List[float]], baseline: str, score: Optional[float]) -> None:
     if score is None:
         print(f"{baseline}: Faithfulness score unavailable for this query.")
@@ -66,7 +64,6 @@ def run_baselines(agent_name: str, queries: List[str]):
 
     return results
 
-
 if __name__ == "__main__":
     # Example usage
     test_queries = [

backend/evaluation/metrics.py

@@ -24,7 +24,7 @@ class MetricsRunner:
         return {
             "faithfulness": faith_res.score,
             "bart_nli": bart_res.score,
-            "factscore": fact_res.score
+            "factscore": fact_res.score,
         }
 
     def extract_faithfulness(self, response: Dict[str, Any]) -> Optional[float]: