Metraipolis › archive › dc_9ff4ebeb97c9
How You Ask Shapes What You Get: A Theory-Seeded Measurement of Articulation in Advice-Seeking LLM Conversations — arXiv:2608.29591
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.29591
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T08:07:03.231Z
- SHA-256 of the archived text
e30e8087ba497a541d552e58ed760abbcc71d599dae3a243eae664c1afc30e5b- Stable URL of this page
https://metraipolis.com/documento/dc_9ff4ebeb97c9
How to cite
How You Ask Shapes What You Get: A Theory-Seeded Measurement of Articulation in Advice-Seeking LLM Conversations — arXiv:2608.29591 — arXiv (en), https://arxiv.org/abs/2608.29591 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_9ff4ebeb97c9 (Metraipolis archive dc_9ff4ebeb97c9, sha256 e30e8087ba497a541d552e58ed760abbcc71d599dae3a243eae664c1afc30e5b)
BibTeX
@misc{dc9ff4ebeb97c9,
title = {How You Ask Shapes What You Get: A Theory-Seeded Measurement of Articulation in Advice-Seeking LLM Conversations — arXiv:2608.29591},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.29591},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_9ff4ebeb97c9; sha256 e30e8087ba497a541d552e58ed760abbcc71d599dae3a243eae664c1afc30e5b},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.29591. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Users articulate the same advice-seeking request in different ways: some specify detailed constraints, others gesture at a vague need. Prior work treats this variation as noise to be averaged away; we instead treat it as a stable, measurable structure in the input distribution. We ask whether articulation (how people ask) forms latent dimensions separable from topic (what they ask about), and whether it is associated with how language models respond. We extract interpretable features from 16,447 advice-seeking prompts pooled from public chat corpora (WildChat, LMSYS, and ShareChat) and recover a small set of latent articulation factors that replicate across train/test splits and across corpora. Because this structure is largely separable from topic, the populations it defines cut across topics and stay invisible to topic- or task-based evaluation. The factors define a handful of recurring articulation styles, one of which stands out: a long-form but information-poor style, roughly one in six prompts in the largest corpus, where models return shorter, vaguer answers and do not ask for clarification even though under-specification is exactly the condition that warrants it. The contrast holds within every topic group and length quintile, and is not under-specification alone -- a second, equally under-specified style does draw clarifying questions. Two independent human annotators reproduce this contrast. We argue that benchmarks should stratify on articulation, and we offer the extracted structure as a measurement instrument for doing so. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-01 in cs.CL). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2608.29591 — 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 2026-10-17T07:47:52.713Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).