Metraipolis › archive › dc_14bfea65010c
Listen to the Latents: Self-Correcting Speech Recognition in Large Audio Language Models Through Hidden-State Interactions — arXiv:2609.02940
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.02940
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T01:11:40.948Z
- SHA-256 of the archived text
be5029399bd1374285d064685a97e76adbeb2f12b10a9d04b66961d72cf88315- Stable URL of this page
https://metraipolis.com/documento/dc_14bfea65010c
How to cite
Listen to the Latents: Self-Correcting Speech Recognition in Large Audio Language Models Through Hidden-State Interactions — arXiv:2609.02940 — arXiv (en), https://arxiv.org/abs/2609.02940 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_14bfea65010c (Metraipolis archive dc_14bfea65010c, sha256 be5029399bd1374285d064685a97e76adbeb2f12b10a9d04b66961d72cf88315)
BibTeX
@misc{dc14bfea65010c,
title = {Listen to the Latents: Self-Correcting Speech Recognition in Large Audio Language Models Through Hidden-State Interactions — arXiv:2609.02940},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.02940},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_14bfea65010c; sha256 be5029399bd1374285d064685a97e76adbeb2f12b10a9d04b66961d72cf88315},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.02940. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Recent automatic speech recognition (ASR) systems increasingly integrate large language models (LLMs) to leverage their semantic knowledge, either externally through logit fusion or internally through warm initialization. However, how to effectively combine these two strategies remains underexplored. In this work, we refine warm-initialized LLM-based ASR models by leveraging their own pre-adaptation base LLMs, focusing on LoRA-adapted settings where the base LLM is preserved. To achieve this, we propose Hybrid Search, a targeted correction strategy motivated by two observations. First, interaction features that characterize the relationship between LLM-based ASR hidden states and base-LLM hidden states provide informative signals about a token's degree of semantic dependence. Second, selectively refining targeted tokens with high semantic dependence improves ASR performance far beyond naive global LLM-correction methods including rescoring and late fusion. Our analysis suggests that, even after semantic knowledge transfer through warm initialization, LLM-based ASR models can still leverage their base LLM to further improve inference-time performance. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-04 in cs.AI). Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.02940 — 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 2026-10-17T00:28:02.928Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-04?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).