Metraipolis › archive › dc_88eb19726bf5
Causal Analysis and Mitigation of Spurious Onsets in Full-Duplex Speech LLMs — arXiv:2609.13445
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.13445
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T23:13:24.901Z
- SHA-256 of the archived text
36a0e4ec01ad75ec8fea69b5bc638a5064baa876bc1d6e0a22df8def9c782a07- Stable URL of this page
https://metraipolis.com/documento/dc_88eb19726bf5
How to cite
Causal Analysis and Mitigation of Spurious Onsets in Full-Duplex Speech LLMs — arXiv:2609.13445 — arXiv (en), https://arxiv.org/abs/2609.13445 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_88eb19726bf5 (Metraipolis archive dc_88eb19726bf5, sha256 36a0e4ec01ad75ec8fea69b5bc638a5064baa876bc1d6e0a22df8def9c782a07)
BibTeX
@misc{dc88eb19726bf5,
title = {Causal Analysis and Mitigation of Spurious Onsets in Full-Duplex Speech LLMs — arXiv:2609.13445},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.13445},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_88eb19726bf5; sha256 36a0e4ec01ad75ec8fea69b5bc638a5064baa876bc1d6e0a22df8def9c782a07},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.13445. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Speech-to-speech LLMs like Moshi, and its derivative PersonaPlex, can listen and speak concurrently through full-duplex generation. However, they can begin speaking inappropriately during prolonged user silence: under digital-zero input, Moshi and PersonaPlex initiate speech in 30% and 27.5% of five-minute continuations, respectively. What causes this spurious speech? We investigate two hypotheses: either repeated sampling selects speech despite persistently low onset probabilities, or self-conditioning on nonspeech outputs causes an abrupt spike in onset probability. We find that, at every observed onset, speech probability spikes by over nine orders of magnitude in one 80-ms frame, supporting the latter hypothesis. Then, to suppress these onsets without blocking genuine responses, we ask a causal counterfactual question: is the model responding to user speech, or would its next-token distribution remain similar if the preceding user input were muted? Accordingly, we suppress onsets whose distributions change little under this intervention. Under realistic microphone noise, our method suppresses spurious onsets, while preserving genuine responses: one-sided 95% lower confidence bounds are 98.68% and 98.82% for Moshi, and 96.90% and 99.25% for PersonaPlex. Our inference-time method runs in real-time without retraining, with 95th-percentile decision time below 61 ms, within the 80-ms frame budget. Our code is available at this https URL . Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-15 in cs.CL). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2609.13445 — 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 2026-10-15T22:51:09.819Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).