Metraipolis › archive › dc_4cd53b3300a3

Beyond Hard Labels: LLM Probabilities Improve Human-Assisted Regression — arXiv:2611.11111

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.11111
License (as archived): CC0-1.0
Archived: 2026-11-10T18:44:35.095Z
SHA-256 of the archived text: e2e07910375113d060a0401c2069c3e0285b8e75a5b940035078554e3ca969de
Stable URL of this page: https://metraipolis.com/documento/dc_4cd53b3300a3

How to cite

Beyond Hard Labels: LLM Probabilities Improve Human-Assisted Regression — arXiv:2611.11111 — arXiv (en), https://arxiv.org/abs/2611.11111 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4cd53b3300a3 (Metraipolis archive dc_4cd53b3300a3, sha256 e2e07910375113d060a0401c2069c3e0285b8e75a5b940035078554e3ca969de)

BibTeX

@misc{dc4cd53b3300a3,
  title = {Beyond Hard Labels: LLM Probabilities Improve Human-Assisted Regression — arXiv:2611.11111},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.11111},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4cd53b3300a3; sha256 e2e07910375113d060a0401c2069c3e0285b8e75a5b940035078554e3ca969de},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.11111. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language models (LLMs) are increasingly used for text annotation, but their outputs are often reduced to a single categorical label before downstream analysis. This hardening step discards the model's relative support across categories, which may contain useful information when category boundaries are ambiguous. We study whether retaining the full probability vector improves downstream regression when LLM annotations are combined with a limited amount of human annotation. Using matched soft and hard representations from three LLMs across three social-science text datasets, we compare direct regression and human-assisted correction. Soft inputs are not uniformly better when used directly as outcomes, but after human correction, using the probability vector consistently improves regression efficiency across models and datasets. Analyses of repeated human judgments and controlled simulations further show that these gains arise when probability variation beyond the hard category carries useful information about the human-coded outcome. Our results identify output representation as an important design choice in LLM-assisted measurement and support preserving probability outputs for downstream statistical analysis. Code and reproducibility materials are available at this https URL

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/2611.11111 — 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 2026-11-10T18:37:52.434Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).