cs.CLNov 8, 2026

Beyond Hard Labels: LLM Probabilities Improve Human-Assisted Regression

Authors: Xinyue Wang, Qunhua Li

Organizations: Department of Statistics, The Pennsylvania State University

Abstract

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 https://anonymous.4open.science/r/soft-label-LLM-D201/

Figures & tables

Appendix figures & tables12 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Definitions, Label Names, or Demonstrations? Auditing What Governs LLM Annotation

    Nov 6, 2026Viraj Mishra, Shubham Gupta, Dhruv Gupta +2LLM-Assisted AnnotationLLM Auditing

  2. Can This LLM Annotate My Data? Estimating Zero-Shot LLM Performance

    Oct 16, 2026Tong Lin, Grant Schoenebeck, Sabina TomkinsLanguage Model Performance PredictionLLM-Assisted Annotation

  3. Multi-Perspective LLM Annotations for Valid Analyses in Subjective Tasks

    Mar 22, 2026Navya Mehrotra, Adam Visokay, Kristina GligorićAnnotator DisagreementLLM-Assisted Annotation