Update README.md

README.md

@@ -2,8 +2,7 @@
 pretty_name: Beyond Binary Instrument QA
 language:
 - en
-license: other
-license_name: cc-by-4.0
+license: cc-by-4.0
 task_categories:
 - audio-classification
 tags:
@@ -55,7 +54,7 @@ Every full `metadata.jsonl` record contains:
 - `audio_archive`: the archive filename in the same configuration directory.
 - `audio_member`: the exact member path inside that archive, beginning with `audio/`.
 
-The remaining benchmark fields—including questions, option order, source sample keys, and gold answers—are unchanged from the individual-file release.
+The remaining benchmark fields—including questions, option order, source sample keys, and gold answers—are unchanged from the metadata used in the accompanying experiments.
 
 ## Extracting audio
 
@@ -93,7 +92,9 @@ Audio archives should be downloaded and extracted separately as described above.
 
 ## Provenance
 
-All released audio is derived from OpenMIC-2018. Gold labels use the benchmark's high-confidence OpenMIC relevance annotations. The 30-second long-context and temporal files are concatenations of three OpenMIC-derived 10-second segments constructed for the accompanying paper.
+All released audio is derived from OpenMIC-2018. The OpenMIC-2018 files used in this work were obtained from the Hugging Face dataset repository `https://huggingface.co/datasets/seungheondoh/openmic-2018`. The authoritative OpenMIC-2018 release and license record is available at https://doi.org/10.5281/zenodo.1432913.
+
+Gold labels use the benchmark's high-confidence OpenMIC relevance annotations. The 30-second long-context and temporal files are concatenations of three OpenMIC-derived 10-second segments constructed for the accompanying paper.
 
 ## Integrity
 
@@ -105,23 +106,34 @@ shasum -a 256 -c SHA256SUMS
 
 ## License
 
-This dataset is released under the Creative Commons Attribution 4.0
-International License (CC BY 4.0).
-
-The audio clips and original instrument annotations are derived from
-OpenMIC-2018, which was released by Spotify AB under CC BY 4.0.
+This release is licensed under the Creative Commons Attribution 4.0 International License (CC BY 4.0).
 
-The BBIQA benchmark reorganizes and extends OpenMIC-2018 into five
-instrument-grounding evaluation configurations. Appropriate attribution
-to both OpenMIC-2018 and Beyond Binary Instrument QA is required when
-using this dataset.
+The source audio clips and instrument annotations are derived from OpenMIC-2018, which is made available by Spotify AB under CC BY 4.0. Beyond Binary Instrument QA selects, reorganizes, annotates, and, for the long-context and temporal-localization configurations, concatenates OpenMIC-derived audio segments for benchmark evaluation.
 
-OpenMIC-2018:
-https://github.com/cosmir/openmic-2018
+When redistributing or adapting this dataset, users must retain the applicable attribution, license, and modification notices. For academic use, please cite both the Beyond Binary Instrument QA paper and the OpenMIC-2018 paper.
 
-CC BY 4.0:
-https://creativecommons.org/licenses/by/4.0/
+- OpenMIC-2018 dataset: https://doi.org/10.5281/zenodo.1432913
+- CC BY 4.0: https://creativecommons.org/licenses/by/4.0/
 
 ## Citation
 
-Please cite the accompanying *Beyond Binary Instrument QA* paper and OpenMIC-2018 when using this benchmark. Formal citation metadata should be added when the paper record is finalized.
+Please cite the accompanying Beyond Binary Instrument QA paper and the following OpenMIC-2018 paper when using this benchmark.
+
+```bibtex
+@article{lee2026beyond,
+  title={Beyond Binary Instrument QA: Probing Instrument Grounding in Music Audio-Language Models},
+  author={Lee, Yujun and Shin, Joonhyeok and Kim, Hyoeun and Shim, Kyuhong},
+  journal={arXiv preprint arXiv:2606.31338},
+  year={2026}
+}
+```
+
+```bibtex
+@inproceedings{humphrey2018openmic,
+  title={Openmic-2018: An open data-set for multiple instrument recognition.},
+  author={Humphrey, Eric and Durand, Simon and McFee, Brian},
+  booktitle={ISMIR},
+  pages={438--444},
+  year={2018}
+}
+```