Metraipolis › archive › dc_dae4d7d326dd
EvoAudio: Recursive Self-Improvement for Audio Understanding — arXiv:2609.27389
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.27389
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T10:58:01.744Z
- SHA-256 of the archived text
d280e21cc79e001e730573228b4b4ff98320f6b45e5f16ce493a06507103716c- Stable URL of this page
https://metraipolis.com/documento/dc_dae4d7d326dd
How to cite
EvoAudio: Recursive Self-Improvement for Audio Understanding — arXiv:2609.27389 — arXiv (en), https://arxiv.org/abs/2609.27389 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_dae4d7d326dd (Metraipolis archive dc_dae4d7d326dd, sha256 d280e21cc79e001e730573228b4b4ff98320f6b45e5f16ce493a06507103716c)
BibTeX
@misc{dcdae4d7d326dd,
title = {EvoAudio: Recursive Self-Improvement for Audio Understanding — arXiv:2609.27389},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.27389},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_dae4d7d326dd; sha256 d280e21cc79e001e730573228b4b4ff98320f6b45e5f16ce493a06507103716c},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.27389. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Audio language models understand what is said far better than how it sounds. Closing this gap takes more than data. Detailed acoustic annotation is costly, labels from stronger models inherit their errors and limits, and fixed data cannot adapt as the learner improves. We therefore propose EvoAudio, a recursive self-improvement system for audio understanding. To our knowledge, it is the first to evolve the model, waveforms, questions, and difficulty in one closed loop. EvoAudio uses the current model's performance to set the focus and difficulty of the next training data. A library of audio tools then constructs questions whose answers follow from how the audio was made, providing verifiable supervision without new human annotation. Reinforcement learning updates the model, and validation decides whether it enters the next evolution round. Across 13 rounds, EvoAudio improves five models with different audio encoders and language backbones on MMSU, MMAU-Pro, and MMAR. It achieves the highest average for every backbone, raising overall performance by up to 6.3 points. The improvement unfolds over successive rounds, with each stronger model starting the next round. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-24 in cs.LG). Categorie arXiv: Sound (cs.SD) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.27389 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-08T10:06:16.814Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-24?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).