Metraipolis › archive › dc_d37337a3ecc8
Correlation-Guided Encoder Selection for Multi-Encoder Large Audio-Language Models — arXiv:2609.18041
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.18041
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T13:49:52.253Z
- SHA-256 of the archived text
b3e93bb2c9d22448250423d721cf7c331e1a4a2a2ad7a4355c3d16712b3f2405- Stable URL of this page
https://metraipolis.com/documento/dc_d37337a3ecc8
How to cite
Correlation-Guided Encoder Selection for Multi-Encoder Large Audio-Language Models — arXiv:2609.18041 — arXiv (en), https://arxiv.org/abs/2609.18041 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d37337a3ecc8 (Metraipolis archive dc_d37337a3ecc8, sha256 b3e93bb2c9d22448250423d721cf7c331e1a4a2a2ad7a4355c3d16712b3f2405)
BibTeX
@misc{dcd37337a3ecc8,
title = {Correlation-Guided Encoder Selection for Multi-Encoder Large Audio-Language Models — arXiv:2609.18041},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.18041},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d37337a3ecc8; sha256 b3e93bb2c9d22448250423d721cf7c331e1a4a2a2ad7a4355c3d16712b3f2405},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.18041. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Multi-encoder fusion extends Large Audio-Language Models (LALMs) beyond speech-centric recognition, but selecting encoders via intuition or exhaustive search often introduces redundant representations and inflates an already constrained compute budget. We propose CUES (Correlation-gUided Encoder Selection), a lightweight heuristic that estimates complementarity through task- and category-level Pearson correlations between encoders' performance profiles, scoring a candidate set from single-encoder evaluations alone--without fusion training during selection. Evaluated on the XARES-LLM benchmark with a frozen SmolLM2-135M backbone (LoRA-adapted) via five-fold cross-validation, CUES consistently identifies the same configuration per track from held-out development splits alone, without using test data for selection. For the broad Track~A suite, CUES selects a cross-family trio (Whisper-medium, mHuBERT-147, and Dasheng-base), achieving a 4.3% relative gain over Whisper-medium (0.771 vs. 0.739). For Track~B text generation, it re-anchors on a focused, speech-only pair (mHuBERT-147 and WavLM-base-plus) and actively abstains from adding a divergent encoder, outperforming mHuBERT-147 by 6.3% (0.589 vs. 0.554). Rather than a failure to scale, this divergence is consistent with a diversity--interference trade-off that CUES navigates per track from correlation signals alone: across the evaluated pool, added cross-family diversity tends toward an inverted-U on broad audio tasks but toward steady degradation on text generation, which favors a focused, speech-anchored set. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-17 in cs.CL). Categorie arXiv: Audio and Speech Processing (eess.AS) ; Computation and Language (cs.CL); Sound (cs.SD). Pagina del lavoro: https://arxiv.org/abs/2609.18041 — 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 2026-10-15T13:48:24.402Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).