Metraipolis › archive › dc_0856f9783e19

Duplex-MPE: Benchmarking Multi-Party Interaction in Full-Duplex Dialogue — arXiv:2609.31948

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.31948
License (as archived)
CC0-1.0
Archived
2026-10-08T01:37:21.166Z
SHA-256 of the archived text
fe2747388b9a50b59bbf795d3dd786e94a66e73a03f9f4c5acc0ba1bfe01a1c7
Stable URL of this page
https://metraipolis.com/documento/dc_0856f9783e19

How to cite

Duplex-MPE: Benchmarking Multi-Party Interaction in Full-Duplex Dialogue — arXiv:2609.31948 — arXiv (en), https://arxiv.org/abs/2609.31948 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0856f9783e19 (Metraipolis archive dc_0856f9783e19, sha256 fe2747388b9a50b59bbf795d3dd786e94a66e73a03f9f4c5acc0ba1bfe01a1c7)
BibTeX
@misc{dc0856f9783e19,
  title = {Duplex-MPE: Benchmarking Multi-Party Interaction in Full-Duplex Dialogue — arXiv:2609.31948},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.31948},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0856f9783e19; sha256 fe2747388b9a50b59bbf795d3dd786e94a66e73a03f9f4c5acc0ba1bfe01a1c7},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.31948. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Real-time full-duplex speech models can listen while speaking, enabling natural interaction without rigid turn boundaries. Existing benchmarks evaluate turn-taking, interruption handling and multi-round dialogue, but largely centre on a designated user rather than an assistant participating in a shared conversation among several people. We introduce Duplex-MPE to evaluate when such an assistant should answer, remain silent or stop speaking. The benchmark contains 2,000 scenarios with three or four human speakers and one assistant, each paired across explicit and implicit addressing of the same request. Models receive continuous conversation audio without transcripts or supplied turn boundaries. Four scores measure fresh response initiation, answer accuracy, silence preservation and stopping when a human resolves a request. We evaluate five open-weight speech systems: MiniCPM-o 4.5, Moshi, FLM-Audio, Voila and Freeze-Omni. MiniCPM-o 4.5 leads on three scored capabilities, while frequent speech from other systems can coexist with inaccurate answers or failures to remain silent. A transcript-based Gemini 3.1 Pro reference responds 64.3 percentage points more often to explicit than implicit requests; paired tests detect no significant response-rate difference for the speech systems.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in eess.AS).
Categorie arXiv: Sound (cs.SD) ; Audio and Speech Processing (eess.AS).
Pagina del lavoro: https://arxiv.org/abs/2609.31948 — 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 2026-10-08T01:11:22.079Z dalla pagina https://arxiv.org/catchup/eess.AS/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).