Metraipolis › archive › dc_e1c24ac01ffa

Hey, guess what: CUES lead to inferences about typicality — arXiv:2610.33239

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.33239
License (as archived)
CC0-1.0
Archived
2026-10-27T08:10:41.656Z
SHA-256 of the archived text
586a551f14b39b51dcdc25370055051cf07cb14671e3192b5903c3adbad9ca42
Stable URL of this page
https://metraipolis.com/documento/dc_e1c24ac01ffa

How to cite

Hey, guess what: CUES lead to inferences about typicality — arXiv:2610.33239 — arXiv (en), https://arxiv.org/abs/2610.33239 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e1c24ac01ffa (Metraipolis archive dc_e1c24ac01ffa, sha256 586a551f14b39b51dcdc25370055051cf07cb14671e3192b5903c3adbad9ca42)
BibTeX
@misc{dce1c24ac01ffa,
  title = {Hey, guess what: CUES lead to inferences about typicality — arXiv:2610.33239},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.33239},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e1c24ac01ffa; sha256 586a551f14b39b51dcdc25370055051cf07cb14671e3192b5903c3adbad9ca42},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.33239. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Throughout life, we accumulate knowledge of how typical certain events are. We use this knowledge in daily conversation, often interpreting meaning beyond what has been said explicitly. In this work, we take steps to better understand this meaning recovery, investigating both atypicality and typicality inferences. First, we construct two corpora, EventTyp and the human-validated EventTypGold. We conduct large-scale human and LLM experiments and find that for humans, three CUES predict inferences about typicality: Commonsense Knowledge, Utterance Emphasis, and Snark. However, in most experimental settings, we find that LLMs lack the prerequisite Commonsense Knowledge necessary to accurately assess an event's general typicality, and therefore struggle to follow these CUES.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.33239 — 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 2026-10-27T07:56:00.186Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).