Metraipolis › archive › dc_cd634441b36b
EMODE: Dynamic Para-Semantic Experts for Emotion-Aware Speech Language Modeling — arXiv:2610.06956
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.06956
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T11:22:57.843Z
- SHA-256 of the archived text
2378dfd2667b30e5c0f7f9e84dd8bdb07aa128fd1501e1a1a457e0c566b2afb3- Stable URL of this page
https://metraipolis.com/documento/dc_cd634441b36b
How to cite
EMODE: Dynamic Para-Semantic Experts for Emotion-Aware Speech Language Modeling — arXiv:2610.06956 — arXiv (en), https://arxiv.org/abs/2610.06956 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cd634441b36b (Metraipolis archive dc_cd634441b36b, sha256 2378dfd2667b30e5c0f7f9e84dd8bdb07aa128fd1501e1a1a457e0c566b2afb3)
BibTeX
@misc{dccd634441b36b,
title = {EMODE: Dynamic Para-Semantic Experts for Emotion-Aware Speech Language Modeling — arXiv:2610.06956},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.06956},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cd634441b36b; sha256 2378dfd2667b30e5c0f7f9e84dd8bdb07aa128fd1501e1a1a457e0c566b2afb3},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.06956. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Large speech language models have demonstrated strong capabilities in unified cross-modal understanding and generation, yet paralinguistic cues, especially emotion, remain difficult to preserve. Existing systems typically rely on entangled acoustic representations, which allow the underlying language model to depend excessively on recovered lexical content instead of grounding its behavior in acoustic-prosodic evidence. We address this limitation with EMODE, an emotion-aware speech language model built around \textbf{Dynamic Para-Semantic Experts (DPSE)}. DPSE decomposes continuous speech features into semantic and paralinguistic pathways, routes them dynamically, and fuses them before integration into the language model. To turn this structural decomposition into functional specialization, EMODE is trained with a three-stage curriculum consisting of semantic warm-up, paralinguistic activation, and joint refinement, guided by Orthogonal Expert Guidance (OEG), Semantic-to-Acoustic Alignment (SAA), and Gating Diversity Regularization (GDR). Experiments on SER test, empathetic response evaluation, and the newly constructed bilingual MEPA benchmark show that EMODE improves the balance between lexical fidelity and emotional sensitivity, strengthens affect-grounded response generation, and exposes the value of explicit para-semantic factorization for robust cross-corpus emotion understanding.
Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL) ; Multimedia (cs.MM); Sound (cs.SD).
Pagina del lavoro: https://arxiv.org/abs/2610.06956 — 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 2026-10-07T11:15:39.406Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).