update the paper link and other info

README.md

@@ -10,7 +10,14 @@ tags:
 ---
 # εar-VAE2
 
-[[Demo Page](https://eps-acoustic-revolution-lab.github.io/EAR_VAE2/)] - [[Paper (coming soon)](#)] - [[Codebase](https://github.com/Eps-Acoustic-Revolution-Lab/EAR_VAE2)]
+**Fourier is Frontier: Frequency-Aware Autoencoding for High-Fidelity Music Reconstruction**
+
+Kangdi Wang<sup>1</sup> · Yusheng Dai<sup>2</sup> · Jin Xu<sup>1†</sup>
+
+<sup>1</sup> Qwen Team, Alibaba &nbsp;&nbsp; <sup>2</sup> Monash University &nbsp;&nbsp; <sup>†</sup> Corresponding author
+
+
+[[Demo Page](https://eps-acoustic-revolution-lab.github.io/EAR_VAE2/)] - [[Paper](https://arxiv.org/abs/2608.19843)] - [[Codebase](https://github.com/Eps-Acoustic-Revolution-Lab/EAR_VAE2)]
 
 ---
 
@@ -43,7 +50,7 @@ Reconstruction quality on **Song Describer Dataset** (546 full tracks, 48 kHz st
 | **εar-VAE2 (base)** | 10.9 | 0.916 | 0.572 | 0.966 |
 | **εar-VAE2 (full)** | **11.3** | **0.870** | **0.461** | **0.973** |
 
-> εar-VAE2 (full) achieves the best spectral fidelity (STFT Dist, Mel Dist) among all systems while matching the phase coherence (CCPC) of the larger εar-VAE baseline.
+> εar-VAE2 (full) achieves the best spectral fidelity (STFT Dist, Mel Dist) among all systems while matching the phase coherence (CCPC) of the εar-VAE baseline.
 
 ## Spec-SnakeBeta
 
@@ -108,36 +115,6 @@ reconstructed = reconstructed[:, :, :orig_len]
 python inference.py --checkpoint ear_vae2.pt --config configs/ear_vae2.json --input input.wav --output output.wav
 ```
 
-<details>
-<summary><b>📁 Project Structure</b></summary>
-
-```
-EAR_VAE2/
-├── README.md
-├── LICENSE
-├── requirements.txt
-├── inference.py                 # CLI inference script
-├── configs/
-│   └── ear_vae2.json           # Full model config (with refiner)
-├── ear_vae2/
-│   ├── __init__.py
-│   └── model.py                # Self-contained EarVAE2 model
-│                               #   (encoder, decoder, Spec-SnakeBeta,
-│                               #    refiner, STFT helpers)
-├── assets/
-│   ├── architecture.png
-│   ├── spec_snakebeta.png
-│   └── input_repr.png
-└── docs/                       # Demo page (GitHub Pages, served from /docs)
-    ├── index.html
-    ├── config.js
-    ├── css/
-    ├── js/
-    ├── assets/
-    └── cases/
-```
-
-</details>
 
 ## Model Details
 
@@ -157,11 +134,14 @@ EAR_VAE2/
 ## Citation
 
 ```bibtex
-@inproceedings{earvae2,
-  title     = {Frequency-Aware Spectral Autoencoding for High-Fidelity Music Reconstruction},
-  author    = {Anonymous},
-  booktitle = {International Conference on Learning Representations (ICLR)},
-  year      = {2026}
+@misc{earvae2,
+  title         = {Fourier is Frontier: Frequency-Aware Autoencoding for High-Fidelity Music Reconstruction},
+  author        = {Kangdi Wang and Yusheng Dai and Jin Xu},
+  year          = {2026},
+  eprint        = {2608.19843},
+  archivePrefix = {arXiv},
+  primaryClass  = {cs.SD},
+  url           = {https://arxiv.org/abs/2608.19843}
 }
 ```
 
@@ -179,6 +159,4 @@ We gratefully acknowledge the following projects that inspired components of εa
 
 ## License
 
-This project is licensed under the [Apache License 2.0](LICENSE).
-
-Copyright 2026 Epsilon Acoustic Revolution Lab.
+This project is licensed under the [Apache License 2.0](LICENSE).