This paper describes our submission to the WMT 2026 Multilingual Instruction Shared Task. We adapt Qwen3-8B with LoRA using organiser-provided data, EuroAlpaca, a multilingual instruction dataset created through task-preserving localisation, and CrossEuroAlpaca, a cross-lingual augmentation that assigns different sourc...
Aleix Sant, Jordi Luque, Carlos Escolano· 0 citations
A voice-enabled smart-home assistant that acts as a household memory by recording conversations and retrieving information from stored and newly captured speech that runs on an Android tablet with a modular architecture.
Aleix Sant, Jordi Luque, Fernando López et al.· 0 citations
A comprehensive benchmark of Federated Learning for multilingual Automatic Speech Recognition (ASR) for multilingual Speech-LLM architectures, evaluating four Speech-LLM architectures on the Multilingual LibriSpeech dataset shows that optimized learning rates are critical for performance.
Jordi Luque, Aleix Sant, Fernando López· 0 citations
This work empirically diagnose the root cause across six per-layer methods and three speech-LLM architectures and proposes a two-pool allocation that normalises encoder and LLM parameters into independent pools, and shows that joint $\ell_2$ sensitivity and the original $(\varepsilon,\delta)$-DP guarantee are unchanged...
Jordi Luque, Fernando López, Aleix Sant· 0 citations
EuroAlpaca is introduced, a task-preserving localisation pipeline and near-parallel resource covering 50 European languages and regional varieties, together with European-IFEval, a multilingual benchmark for verifiable instruction following, showing that preserving task semantics is essential for multilingual instructi...
Aleix Sant, Jordi Luque, Carlos Escolano· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
