Upload folder using huggingface_hub

.gitattributes

@@ -34,3 +34,14 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 methodology.png filter=lfs diff=lfs merge=lfs -text
+docs/architecture_diagrams/methodology.png filter=lfs diff=lfs merge=lfs -text
+docs/visualizations/COCO_DF_C110B00000_00539519_vaas.png filter=lfs diff=lfs merge=lfs -text
+docs/visualizations/COCO_DF_E000B00011_00834527_vaas.png filter=lfs diff=lfs merge=lfs -text
+docs/visualizations/COCO_DF_E000B04010_00863031_vaas.png filter=lfs diff=lfs merge=lfs -text
+docs/visualizations/COCO_DF_E000B04011_00898719_vaas.png filter=lfs diff=lfs merge=lfs -text
+docs/visualizations/COCO_DF_E000B04016_00730443_vaas.png filter=lfs diff=lfs merge=lfs -text
+docs/visualizations/COCO_DF_E000B60100_00731826_vaas.png filter=lfs diff=lfs merge=lfs -text
+docs/visualizations/COCO_DF_I000B00000_00966250_vaas.png filter=lfs diff=lfs merge=lfs -text
+docs/visualizations/COCO_DF_S000B00000_00058274_vaas.png filter=lfs diff=lfs merge=lfs -text
+docs/visualizations/COCO_DF_S000B00000_00120651_vaas.png filter=lfs diff=lfs merge=lfs -text
+docs/visualizations/COCO_DF_S000B00000_00251471_vaas.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -21,12 +21,15 @@ It combines **global attention-based reasoning** with **patch-level self-consist
 
 Rather than making binary decisions, VAAS estimates **where anomalies occur** and **how strongly they deviate from learned visual regularities**, enabling explainable image analysis and integrity assessment.
 
-Paper link: *to be added upon publication*
+## Read Paper
+
+- [Arxiv version](https://arxiv.org/abs/2512.15512)
+- [Conference version](https://arxiv.org/abs/2512.15512)
 ---
 
 ## Architecture Overview
 
-![VAAS Methodology](./methodology.png)
+![VAAS Methodology](docs/architecture_diagrams/methodology.png)
 
 VAAS consists of two complementary components:
 
@@ -44,22 +47,6 @@ These components are combined via a hybrid scoring mechanism:
 
 `S_H` provides a continuous measure of anomaly intensity rather than a binary decision.
 
----
-
-## Model Variant
-
-This release corresponds to:
-
-- **VAAS v1**
-- Trained on **10% of the DF2023 dataset**
-- Input resolution: `224 × 224`
-- Outputs:
-  - Global anomaly score (`S_H`)
-  - Component scores (`S_F`, `S_P`)
-  - Dense anomaly map (`224 × 224`)
-
-Future releases will scale training data size, include cross-dataset evaluation, and explore model compression.
-
 ---
 ## Intended Use
 
@@ -124,7 +111,7 @@ pipeline = VAASPipeline.from_pretrained(
 image = Image.open("example.jpg").convert("RGB")
 result = pipeline(image)
 
-print(result["S_H"])
+print(result)
 anomaly_map = result["anomaly_map"]
 ```
 
@@ -163,7 +150,26 @@ This will save a figure containing:
 * Global attention overlays
 * A gauge-style visualization of the hybrid anomaly score
 
-![Inference with visual example](./methodology.png)
+For examples:
+
+![Inference with visual example](docs/visualizations/COCO_DF_I000B00000_00966250_vaas.png)
+![Inference with visual example](docs/visualizations/COCO_DF_S000B00000_00120651_vaas.png)
+![Inference with visual example](docs/visualizations/COCO_DF_C110B00000_00539519_vaas.png)
+---
+
+## Model Variant
+
+This release corresponds to:
+
+- **VAAS v1**
+- Trained on **10% of the DF2023 dataset**
+- Input resolution: `224 × 224`
+- Outputs:
+  - Global anomaly score (`S_H`)
+  - Component scores (`S_F`, `S_P`)
+  - Dense anomaly map (`224 × 224`)
+
+Future releases will scale training data size, include cross-dataset evaluation, and explore model compression.
 
 ---