Metraipolis › archive › dc_43b11ed91c3b

Federated Multilingual Speech-LLMs: Architecture and Aggregation Strategy Benchmarking — arXiv:2609.23825

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.23825
License (as archived)
CC0-1.0
Archived
2026-10-09T06:49:56.127Z
SHA-256 of the archived text
dd62c50b3ee44f04a5bfaecc7e523c260af2c77452156f95e27da0eb993865f9
Stable URL of this page
https://metraipolis.com/documento/dc_43b11ed91c3b

How to cite

Federated Multilingual Speech-LLMs: Architecture and Aggregation Strategy Benchmarking — arXiv:2609.23825 — arXiv (en), https://arxiv.org/abs/2609.23825 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_43b11ed91c3b (Metraipolis archive dc_43b11ed91c3b, sha256 dd62c50b3ee44f04a5bfaecc7e523c260af2c77452156f95e27da0eb993865f9)
BibTeX
@misc{dc43b11ed91c3b,
  title = {Federated Multilingual Speech-LLMs: Architecture and Aggregation Strategy Benchmarking — arXiv:2609.23825},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.23825},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_43b11ed91c3b; sha256 dd62c50b3ee44f04a5bfaecc7e523c260af2c77452156f95e27da0eb993865f9},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.23825. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We present a comprehensive benchmark of Federated Learning (FL) for multilingual Automatic Speech Recognition (ASR), evaluating four Speech-LLM architectures on the Multilingual LibriSpeech dataset. We compare FedAvg and FedProx across frozen and unfrozen encoder configurations, demonstrating that optimized learning rates are critical for performance. Specifically, independently tuning the learning rates for the speech encoder, connector, and decoder yields the lowest error rates, with full three-component adaptation (LoRA for encoder and decoder, full training for the connector) producing the best FL results. We observe that FedProx efficacy is architecture-dependent, providing notable advantages in multilingual pre-trained architectures (e.g., EuroLLM over TinyLlama when keeping the encoder fixed); this indicates that LLM backbone capacity plays a key role in mediating resilience to heterogeneous data distributions. These findings offer concrete design guidance for deploying multilingual Speech-LLMs in privacy-sensitive, distributed environments.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-22 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.23825 — 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 2026-10-09T06:48:45.009Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-22?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).