Upload model card, weights, and evaluation artifacts

README.md

@@ -32,20 +32,44 @@ This model is designed to detect question regions in exam booklets, worksheets,
 
 This release checkpoint was trained for 50 epochs on the cleaned one-class question detection dataset.
 
+Compared with the previously published HF YOLO11M baseline, this release improves held-out mAP50-95 by `+0.026` on the old test set and `+0.062` on the combined test set.
+
 ## Test results
 
 ### Old held-out test set
 
 | Model | Precision | Recall | mAP50 | mAP50-95 |
 | --- | --- | --- | --- | --- |
+| HF YOLO11M | 0.962 | 0.980 | 0.979 | 0.898 |
+| YOLO26M Question Segmentation | 0.990 | 0.982 | 0.988 | 0.920 |
 | YOLO26N Question Segmentation | 0.988 | 0.983 | 0.989 | 0.924 |
 
 ### Combined held-out test set
 
 | Model | Precision | Recall | mAP50 | mAP50-95 |
 | --- | --- | --- | --- | --- |
+| HF YOLO11M | 0.962 | 0.973 | 0.984 | 0.900 |
+| YOLO26M Question Segmentation | 0.987 | 0.978 | 0.992 | 0.957 |
 | YOLO26N Question Segmentation | 0.991 | 0.988 | 0.993 | 0.962 |
 
+### Benchmark graphs
+
+![Held-out mAP50-95](benchmark_map50_95.png)
+![Held-out precision and recall](benchmark_precision_recall.png)
+
+## Qualitative examples
+
+The newer YOLO26 checkpoints improve most clearly in two cases:
+
+- Less false positives on pages with no questions.
+- Better recovery of harder placements such as one-row questions and questions that occupy an entire column.
+
+Example page with no saved questions: GT + HF YOLO11M + YOLO26M + YOLO26N.
+
+![No-question page comparison](comparison_examples/10__sourcepdf__yks_tyt_2025_kitapcik_d250__page_1.png)
+
+Additional side-by-side examples are included in `comparison_examples/`.
+
 ## Usage
 
 ```python