Metraipolis › archive › dc_76c296053afa

What and Where in Identifying and Quantifying Conventions in Humans and AI Agents — arXiv:2610.20205

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.20205
License (as archived)
CC0-1.0
Archived
2026-10-15T13:50:46.770Z
SHA-256 of the archived text
99cac7b2fdff2cacb0d233fa612e4859d5495a9575fa3c6210df9372bd960b2f
Stable URL of this page
https://metraipolis.com/documento/dc_76c296053afa

How to cite

What and Where in Identifying and Quantifying Conventions in Humans and AI Agents — arXiv:2610.20205 — arXiv (en), https://arxiv.org/abs/2610.20205 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_76c296053afa (Metraipolis archive dc_76c296053afa, sha256 99cac7b2fdff2cacb0d233fa612e4859d5495a9575fa3c6210df9372bd960b2f)
BibTeX
@misc{dc76c296053afa,
  title = {What and Where in Identifying and Quantifying Conventions in Humans and AI Agents — arXiv:2610.20205},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.20205},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_76c296053afa; sha256 99cac7b2fdff2cacb0d233fa612e4859d5495a9575fa3c6210df9372bd960b2f},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.20205. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Cooperative success rests in part on implicit communication. Partners converge on conventions: regularities that give an action meaning beyond its direct effect. Yet conventions are usually observed only indirectly: a score drop when partners are swapped shows that a convention existed, but not what it was or where it was used. In the card game Hanabi, the exact probability that a play fails can be computed from the hints received and the cards in view. Players who use conventions succeed more often than this prediction allows; the resulting margin is the convention gap. Off-belief learning provides a hierarchy of agents (levels 1-5) whose first level uses no conventions by construction and whose later levels use progressively more. For humans and these agents on one scale, we ask what the conventions are and where they are used. On what: in two independently collected corpora of human games, humans show a larger gap than every agent level. On the conventions both populations use, however, humans do no better than level 2. The human advantage is concentrated in a situation that no agent level learns to use: hints that point to a card other than the ones they touch. On where: conventions depend on what the player already knows about the card, not on how recently it was hinted, and they are used most in the mid-game; humans use them earlier and later in the game than the agents. In interventions on the agents, blocking a single high-gap situation costs little, because meaning reroutes to other situations, but blocking several together costs far more than the sum of the single costs. Thus, the convention gap turns conventions from an inferred cause of score loss into a measurable, comparable, and causally testable object.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.MA, nuovi lavori del giorno).
Categorie arXiv: Human-Computer Interaction (cs.HC) ; Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA).
Pagina del lavoro: https://arxiv.org/abs/2610.20205 — 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 2026-10-15T13:48:35.464Z dalla pagina https://arxiv.org/list/cs.MA/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).