Metraipolis › archive › dc_4cdebab2e433

Speak for Me: Giving LLMs the Situational Awareness to Participate in a Meeting — arXiv:2609.03923

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.03923
License (as archived): CC0-1.0
Archived: 2026-10-17T01:11:27.617Z
SHA-256 of the archived text: 4e1f69ea4af5918a5e4a4cdb8ed635343f17949019a71581b5316fb52126f2e8
Stable URL of this page: https://metraipolis.com/documento/dc_4cdebab2e433

How to cite

Speak for Me: Giving LLMs the Situational Awareness to Participate in a Meeting — arXiv:2609.03923 — arXiv (en), https://arxiv.org/abs/2609.03923 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4cdebab2e433 (Metraipolis archive dc_4cdebab2e433, sha256 4e1f69ea4af5918a5e4a4cdb8ed635343f17949019a71581b5316fb52126f2e8)

BibTeX

@misc{dc4cdebab2e433,
  title = {Speak for Me: Giving LLMs the Situational Awareness to Participate in a Meeting — arXiv:2609.03923},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.03923},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4cdebab2e433; sha256 4e1f69ea4af5918a5e4a4cdb8ed635343f17949019a71581b5316fb52126f2e8},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.03923. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

In online meeting delegation, LLM agents fail to recognize when to speak. With no structured way to track stances, coverage, and floor, they miss the moments where they should contribute. Prompt-only delegates stay silent on 51.4% of the absent participant's talking opportunities on the AMI corpus. We present CAPA (Collaborative Agent Predictive Architecture), an architecture for online meeting delegation. A Perceiver updates the meeting state from each observed turn. A Predictor forecasts how the conversation will continue. A Controller decides whether to speak and which proposition to surface. A Generator phrases the chosen contribution in the participant's style. Two judges score the forecast and the action against the next observed turn. A Recalibrator updates the meeting state from those verdicts for future decisions. To evaluate online delegation, we introduce an episode-level protocol that scores whether, when, and what a delegate contributes around the participant's actual idea units. The protocol's schema-constrained LLM judges align with human annotations at Cohen's kappa = 0.71. On 137 AMI meetings, CAPA reduces the silence rate from 51.4% to 2.5%, doubles credited recovery (26.1 --> 52.2), and keeps hallucination at 0.6%. The failure mode shifts from omission to selection, with each residual near-miss attributable to a specific module of the architecture. Mechanism ablations identify the meeting state as the lever that closes the recognition gap, where raw-context scaling alone does not.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-04 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.03923 — 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 2026-10-17T00:28:02.892Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-04?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).