Upload TrailRAG cross-encoder model for msmarco

CECorrelationEvaluator_validation_results.csv

@@ -1,19 +1,7 @@
 epoch,steps,Pearson_Correlation,Spearman_Correlation
-0,-1,nan,nan
-1,-1,nan,nan
-2,-1,nan,nan
-3,-1,nan,nan
-4,-1,nan,nan
-5,-1,nan,nan
-0,-1,0.8234889274002977,0.8416879460431413
-1,-1,0.8399841527047024,0.8235109696115698
-2,-1,0.900489371141846,0.8978470694884074
-3,-1,0.9201243213920631,0.915655588453799
-4,-1,0.9199197430314849,0.9151963740959492
-5,-1,0.9211060780954796,0.9159751768244645
-0,-1,0.8445642100994721,0.8505106472723202
-1,-1,0.8685314728766171,0.8408928674396655
-2,-1,0.9156971557343321,0.8977678640249434
-3,-1,0.9292048103277807,0.9111684823827604
-4,-1,0.9326061987049302,0.9129828509937034
-5,-1,0.934848195218772,0.915442276818865
+0,-1,0.9065637785941512,0.9096316916400511
+1,-1,0.937816813815745,0.9120291864785336
+2,-1,0.9589037077805673,0.9377012554777174
+3,-1,0.9620169410539339,0.9427208640595239
+4,-1,0.9651100139157162,0.944031141004433
+5,-1,0.9659627628662212,0.9446488429927474

README.md

@@ -21,17 +21,17 @@ model-index:
       type: msmarco
     metrics:
     - type: mse
-      value: 0.0618574036824182
+      value: 0.0423588519082496
     - type: mae
-      value: 0.1473706976051132
+      value: 0.1121706619454281
     - type: rmse
-      value: 0.2487114868324707
+      value: 0.2058126621669562
     - type: r2_score
-      value: 0.588492937027161
+      value: 0.7490766636371498
     - type: pearson_correlation
-      value: 0.857523177971012
+      value: 0.9093360796297332
     - type: spearman_correlation
-      value: 0.8264641527356917
+      value: 0.8886928996060736
 ---
 
 # TrailRAG Cross-Encoder: MS MARCO Enhanced
@@ -53,20 +53,20 @@ This is a fine-tuned cross-encoder model specifically optimized for **Passage Ra
 
 | Metric | Value | Description |
 |--------|-------|-------------|
-| **MSE** | **0.061857** | Mean Squared Error (lower is better) |
-| **MAE** | **0.147371** | Mean Absolute Error (lower is better) |
-| **RMSE** | **0.248711** | Root Mean Squared Error (lower is better) |
-| **R² Score** | **0.588493** | Coefficient of determination (higher is better) |
-| **Pearson Correlation** | **0.857523** | Linear correlation (higher is better) |
-| **Spearman Correlation** | **0.826464** | Rank correlation (higher is better) |
+| **MSE** | **0.042359** | Mean Squared Error (lower is better) |
+| **MAE** | **0.112171** | Mean Absolute Error (lower is better) |
+| **RMSE** | **0.205813** | Root Mean Squared Error (lower is better) |
+| **R² Score** | **0.749077** | Coefficient of determination (higher is better) |
+| **Pearson Correlation** | **0.909336** | Linear correlation (higher is better) |
+| **Spearman Correlation** | **0.888693** | Rank correlation (higher is better) |
 
 ### Training Details
 
 - **Training Duration**: 21 minutes
 - **Epochs**: 6
 - **Early Stopping**: No
-- **Best Correlation Score**: 0.915442
-- **Final MSE**: 0.061857
+- **Best Correlation Score**: 0.944649
+- **Final MSE**: 0.042359
 
 ### Training Configuration
 

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e93dddb368dadda65587276326b1bce15d11fc1aab6831d8512ad3668b5d0c7e
+oid sha256:6e71404ed959da112de9f8b1884ef010fb322eef32651c4f37568e7080079dc8
 size 90866412