Metraipolis › archive › dc_56a8a79ed180
Knowing When Not to Answer: Pseudo-Ensembles for Abstention in Music Audio-Language Models — arXiv:2609.04362
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.04362
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T22:35:51.498Z
- SHA-256 of the archived text
67a90540510be2334924eef17595981d387facb5e1cd9494dc730c9d2653ab10- Stable URL of this page
https://metraipolis.com/documento/dc_56a8a79ed180
How to cite
Knowing When Not to Answer: Pseudo-Ensembles for Abstention in Music Audio-Language Models — arXiv:2609.04362 — arXiv (en), https://arxiv.org/abs/2609.04362 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_56a8a79ed180 (Metraipolis archive dc_56a8a79ed180, sha256 67a90540510be2334924eef17595981d387facb5e1cd9494dc730c9d2653ab10)
BibTeX
@misc{dc56a8a79ed180,
title = {Knowing When Not to Answer: Pseudo-Ensembles for Abstention in Music Audio-Language Models — arXiv:2609.04362},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.04362},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_56a8a79ed180; sha256 67a90540510be2334924eef17595981d387facb5e1cd9494dc730c9d2653ab10},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.04362. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Music audio-language models are evaluated almost entirely by accuracy on multiple-choice questions. This protocol forces the model to commit to an option, so a lucky guess looks the same as real musical understanding. What is missing is a way to tell when the model does not know the answer, so that it can abstain instead of guessing. The usual solution, an ensemble of independently trained models, is far too expensive here, which leaves the entropy of a single predictive distribution as the only available confidence signal. We instead build pseudo-ensembles from one pretrained model by perturbing its input in ways that cannot change the correct answer, then averaging the resulting distributions over the options. Our main construction simply shuffles the order in which the candidate answers are presented; we also study ensembles built from corrupted audio and from swapped option labels. A pseudo-ensemble gives several predictive distributions per question, so it supports the full family of ensemble-based uncertainty measures (entropy of the expected distribution, expected entropy, and their difference, the mutual information) rather than entropy alone. Evaluating TinyMU on MuChoMusic, we find that averaging over four option orderings raises accuracy from 55.7% to 59.2%, and that the resulting uncertainty measures rank the model's errors better than the single-pass entropy baseline, reducing the area under the error retention curve from 0.293 to 0.261. All of this costs a few extra forward passes and no retraining, which makes abstention practical for compact music audio-language models. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-07 in cs.CL). Categorie arXiv: Sound (cs.SD) ; Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2609.04362 — 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 2026-10-16T22:21:16.314Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).