A Rose by Any Other Name: LLM-Generated Explanations Are Good Proxies for Human Explanations to Collect Label Distributions on NLI

Documenti analoghi

• "Seeing the Big through the Small": Can LLMs Approximate Human Judgment Distributions on NLI from a Few Explanations?
  di: Chen, Beiduo, et al.
  Pubblicazione: (2024)
• Agree, Disagree, Explain: Decomposing Human Label Variation in NLI through the Lens of Explanations
  di: Hong, Pingjun, et al.
  Pubblicazione: (2025)
• EVADE: LLM-Based Explanation Generation and Validation for Error Detection in NLI
  di: Zuo, Longfei, et al.
  Pubblicazione: (2025)
• Human Label Variation as Stable Signal: Learning Annotator-Specific Explanation Behavior via Cross-Annotator Preference Optimization
  di: Chen, Beiduo, et al.
  Pubblicazione: (2026)
• LiTEx: A Linguistic Taxonomy of Explanations for Understanding Within-Label Variation in Natural Language Inference
  di: Hong, Pingjun, et al.
  Pubblicazione: (2025)