docs: WACV 2026 not 2027, seven test sets, exact numbers, link the training data

README.md

@@ -20,7 +20,7 @@ OracleZoom drives Chain-of-Zoom's recursive 4x super-resolution out to 256x whil
 
 [![Base](https://img.shields.io/badge/Base-SD3%20%2B%20Qwen2.5--VL-blue)](https://huggingface.co/stabilityai/stable-diffusion-3-medium-diffusers)
 [![Method](https://img.shields.io/badge/Method-Chain--of--Zoom-orange)](https://github.com/bryanswkim/Chain-of-Zoom)
-[![Paper](https://img.shields.io/badge/Paper-WACV%202027%20(in%20submission)-red)](#)
+[![Paper](https://img.shields.io/badge/Paper-WACV%202026%20(in%20submission)-red)](#)
 [![License](https://img.shields.io/badge/License-CC--BY--NC--4.0-lightgrey)](https://creativecommons.org/licenses/by-nc/4.0/)
 
 ## Quickstart (one image, all scales)
@@ -50,6 +50,11 @@ Stable Diffusion 3-medium and Qwen2.5-VL-3B download automatically on first run.
 
 > **Batching many images:** `inference.py` exposes `zoom_image(sr, model, proc, pvi, image_path, out_dir)`. Build the models once (`build_sr(...)`, `build_vlm(...)`) and call `zoom_image` in a loop over your images.
 
+## Training data
+The curated training set is released separately at
+[dipta007/OracleZoom-4KLSDB-train](https://huggingface.co/datasets/dipta007/OracleZoom-4KLSDB-train).
+The released model uses its `1k` config (1,000 curated 4K images).
+
 ## What's in this repo
 | Path | What it is |
 |---|---|
@@ -65,12 +70,12 @@ Recursive SR (Chain-of-Zoom) reuses a 4x backbone step after step to reach 16x-2
 **Privileged-latent distillation.** A *privileged teacher* is shown the ground-truth high-resolution patch **at training time only** and distills its real detail into the blind student, in **decode space**. Only a small adapter is trained; the backbone, VAE, and prompter stay frozen. **A KL leash** to the deployed backbone keeps a deep sharpness reward from drifting into a metric-gaming texture, so detail stays faithful. Trained: rank-16 adapter (7.1M params), 1,000 curated 4K images; beta_reward 0.4, beta_kl 8.0. The released weights have this adapter already merged in.
 
 ## Results
-Under Chain-of-Zoom's exact protocol on a curated 4K benchmark and six test sets:
+Under Chain-of-Zoom's exact protocol on a curated 4K benchmark and seven test sets:
 
 | Axis | Metric | Ours | CoZ / best baseline |
 |---|---|---|---|
-| Sharpness (no-reference) | CLIPIQA @256x | **0.71** | <= 0.58 |
-| Fidelity @4x (ground truth exists) | LPIPS | **0.20** | 0.22 (CoZ) |
+| Sharpness (no-reference) | CLIPIQA @256x | **0.706** | 0.579 (CoZ) |
+| Fidelity @4x (ground truth exists) | LPIPS | **0.199** | 0.215 (CoZ) |
 | Deep faithfulness (MLLM judge, 64-256x) | preferred vs CoZ | **68-78%** | - |
 | Deep faithfulness | hallucination rate vs CoZ | **2-5x lower** | - |
 
@@ -85,11 +90,11 @@ The `coz/` code and the checkpoints in `ckpt/` are from [Chain-of-Zoom](https://
 
 ## Citation
 ```bibtex
-@inproceedings{dipta2027oraclezoom,
+@inproceedings{dipta2026oraclezoom,
   title={OracleZoom: Privileged-Latent Distillation for Faithful Extreme Super-Resolution},
   author={Shubhashis Roy Dipta},
-  year={2027},
-  note={In submission, WACV 2027}
+  year={2026},
+  note={In submission, WACV 2026}
 }
 ```
 Please also cite Chain-of-Zoom and OSEDiff.