De-anonymize and polish README: add ICML 2026 paper info, arXiv DOI, cross-links to vocsim/* repos

README.md

@@ -29,10 +29,10 @@ dataset_info:
     dtype: string
   splits:
   - name: train
-    num_bytes: 5452179735 # ~5.45 GB
+    num_bytes: 5452179735
     num_examples: 114641
-  download_size: 5500616162 # ~5.5 GB
-  dataset_size: 5452179735 # ~5.45 GB
+  download_size: 5500616162
+  dataset_size: 5452179735
 configs:
 - config_name: default
   data_files:
@@ -40,47 +40,68 @@ configs:
     path: data/train-*
 ---
 
-# VocSim: A Training-Free Benchmark for Content Identity in Single-Source Audio Embeddings
-[![GitHub Repository](https://img.shields.io/badge/GitHub-vocsim-black?logo=github)](https://github.com/anonymoussubmission0000/vocsim)
-[![Paper: arXiv](https://img.shields.io/badge/Paper-arXiv%20(TBD)-red.svg)](#)
+# VocSim — Public Benchmark
+
+[![GitHub](https://img.shields.io/badge/GitHub-vocsim%2Fbenchmark-black?logo=github)](https://github.com/vocsim/benchmark)
+[![Leaderboard](https://img.shields.io/badge/%F0%9F%A4%97%20Leaderboard-vocsim%2FVocSim-green)](https://huggingface.co/spaces/vocsim/VocSim)
 [![License: CC BY 4.0](https://img.shields.io/badge/License-CC%20BY%204.0-blue.svg)](https://creativecommons.org/licenses/by/4.0/)
 
-**VocSim** evaluates how well neural audio embeddings generalize for **zero-shot audio similarity**. It tests recognizing fine-grained acoustic similarity without specific similarity training.
+The public split of **VocSim**, a training-free benchmark for zero-shot content identity in single-source audio embeddings. VocSim probes the intrinsic geometric quality of frozen audio representations: do acoustically variable instances of the same content land near each other in embedding space, without any task-specific training?
 
----
+> Basha, M., Zai, A. T., Stoll, S., & Hahnloser, R. H. R. *VocSim: A Training-free Benchmark for Zero-shot Content Identity in Single-source Audio.* ICML 2026. [arXiv:2512.10120](https://doi.org/10.48550/arXiv.2512.10120)
 
-## Key Features
+## What's here
 
-*   **Diverse Sources:** Human speech (phones, words, utterances), birdsong, otter calls, environmental sounds.
-*   **Varied Conditions:** Spans clean to noisy recordings, short (<100ms) to long durations, few to many classes per subset.
-*   **Standardized:** All audio is 16kHz mono.
+- **114,641 clips** across **15 public subsets**, drawn from 19 source corpora.
+- Domains: human speech (phones, words, utterances), animal vocalizations (birdsong, otter calls), environmental sounds.
+- Conditions: clean to noisy, sub-100ms to multi-second, few to thousands of classes per subset.
+- All audio standardized to **16 kHz mono**.
+- Single-source only — no overlapping speakers or simultaneous sources — so evaluation isolates content representation from source separation.
 
----
+Four additional **blind out-of-distribution subsets** (low-resource speech in Shipibo-Conibo and Chintang) are held out for server-side evaluation via the [leaderboard](https://huggingface.co/spaces/vocsim/VocSim).
 
-## Data Format
+## Schema
 
 ```python
 {
-  'audio': {'array': array([...], dtype=float32), 'sampling_rate': 16000},
-  'subset': 'HW1',      # Origin identifier
-  'speaker': 'spk_id',  # Speaker/Animal/Source ID or 'N/A'
-  'label': 'hello'      # Ground truth class for similarity
+  "audio":  {"array": np.ndarray, "sampling_rate": 16000},
+  "subset": "HW1",        # source-corpus tag (see paper for the full list)
+  "speaker": "spk_042",   # speaker / animal / source ID, or "N/A"
+  "label":  "hello",      # ground-truth class for similarity
 }
 ```
-Train split: 114,641 public examples from 15 subsets for evaluation.
 
-Blind Test Sets: 4 additional subsets held out privately.
+## Quick start
+
+```python
+from datasets import load_dataset
+
+ds = load_dataset("vocsim/public", split="train")
+print(ds[0])
+```
+
+For end-to-end evaluation (feature extraction, distance computation, P@k / GSR), use the reference pipeline at [github.com/vocsim/benchmark](https://github.com/vocsim/benchmark).
+
+## Companion datasets
+
+| Dataset | Purpose |
+|---|---|
+| [`vocsim/avian-perception-benchmark`](https://huggingface.co/datasets/vocsim/avian-perception-benchmark) | Alignment of embeddings with zebra-finch perceptual judgments |
+| [`vocsim/mouse-strain-classification-benchmark`](https://huggingface.co/datasets/vocsim/mouse-strain-classification-benchmark) | C57 vs DBA USV classification |
+| [`vocsim/mouse-identity-classification-benchmark`](https://huggingface.co/datasets/vocsim/mouse-identity-classification-benchmark) | Individual-mouse identification from USVs |
+
+## Licensing
+
+Aggregation and metadata are released under CC BY 4.0. Each source corpus retains its original license; see Appendix A.1.1 of the paper for a per-source breakdown.
 
 ## Citation
-```bib
-@inproceedings{vocsim_authors_2025,
-  title={VocSim: A Training-Free Benchmark for Content Identity in Single-Source Audio Embeddings},
-  author={Anonymous Authors},
-  booktitle={Conference/Journal},
-  year={2025},
-  url={[Link to paper upon DOI]}
+
+```bibtex
+@inproceedings{basha2026vocsim,
+  title     = {VocSim: A Training-free Benchmark for Zero-shot Content Identity in Single-source Audio},
+  author    = {Basha, Maris and Zai, Anja T. and Stoll, Sabine and Hahnloser, Richard H. R.},
+  booktitle = {Proceedings of the 43rd International Conference on Machine Learning (ICML)},
+  year      = {2026},
+  doi       = {10.48550/arXiv.2512.10120}
 }
 ```
-## License
-
-CC BY 4.0 - Creative Commons Attribution 4.0 International.