Revised for training

docs/validation_guide.md

@@ -118,28 +118,34 @@ python scripts/comprehensive_validation.py \
 
 **Purpose**: Head-to-head performance comparisons with baseline models.
 
-#### Retriever Comparison (`compare_retriever.py`)
+#### Unified Model Comparison (`compare_models.py`)
 
 ```bash
-python scripts/compare_retriever.py \
-    --finetuned_model_path ./output/bge-m3-enhanced/final_model \
-    --baseline_model_path BAAI/bge-m3 \
+# Compare retriever
+python scripts/compare_models.py \
+    --model_type retriever \
+    --finetuned_model ./output/bge-m3-enhanced/final_model \
+    --baseline_model BAAI/bge-m3 \
     --data_path data/datasets/examples/embedding_data.jsonl \
     --batch_size 16 \
-    --max_samples 1000 \
-    --output ./retriever_comparison.txt
-```
+    --max_samples 1000
 
-#### Reranker Comparison (`compare_reranker.py`)
-
-```bash
-python scripts/compare_reranker.py \
-    --finetuned_model_path ./output/bge-reranker/final_model \
-    --baseline_model_path BAAI/bge-reranker-base \
+# Compare reranker
+python scripts/compare_models.py \
+    --model_type reranker \
+    --finetuned_model ./output/bge-reranker/final_model \
+    --baseline_model BAAI/bge-reranker-base \
     --data_path data/datasets/examples/reranker_data.jsonl \
     --batch_size 16 \
-    --max_samples 1000 \
-    --output ./reranker_comparison.txt
+    --max_samples 1000
+
+# Compare both at once
+python scripts/compare_models.py \
+    --model_type both \
+    --finetuned_retriever ./output/bge-m3-enhanced/final_model \
+    --finetuned_reranker ./output/bge-reranker/final_model \
+    --retriever_data data/datasets/examples/embedding_data.jsonl \
+    --reranker_data data/datasets/examples/reranker_data.jsonl
 ```
 
 **Output**: Tab-separated comparison tables with metrics and performance deltas.

docs/validation_system_guide.md

@@ -0,0 +1 @@
+