Salvato in:
Dettagli Bibliografici
Autori principali: Gao, Lingyu, Monroe, Will, Smith, David, Jemison, Meghan, Lee, Jackie
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:https://arxiv.org/abs/2605.26070
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866918522098024448
author Gao, Lingyu
Monroe, Will
Smith, David
Jemison, Meghan
Lee, Jackie
author_facet Gao, Lingyu
Monroe, Will
Smith, David
Jemison, Meghan
Lee, Jackie
contents Annotating speaker attributes from text is inherently ambiguous, particularly in multilingual settings where demographic and social cues are implicit and culturally variable. We propose a human-large language model (LLM) collaborative re-annotation framework for stabilizing multilingual speaker-attribute labels under practical resource constraints. Starting from a noisy corpus, we use LLMs to surface recurring annotation rationales through iterative interaction with experts, and apply disagreement-focused sampling for targeted re-annotation. Using this framework, we construct WhoSaidIt, a multilingual dataset covering nine speaker-attribute labels. We quantify divergence between original and revised annotations, benchmark recent LLMs, and analyze the effect of explicit rationales on model behavior. Our results reveal substantial cross-lingual differences in annotation decisions and demonstrate both the strengths and limitations of LLMs in speaker-attribute classification.
format Preprint
id arxiv_https___arxiv_org_abs_2605_26070
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle WhoSaidIt: Human-LLM Collaborative Annotation for Text-Based Multilingual Speaker-Attribute Classification
Gao, Lingyu
Monroe, Will
Smith, David
Jemison, Meghan
Lee, Jackie
Computation and Language
Annotating speaker attributes from text is inherently ambiguous, particularly in multilingual settings where demographic and social cues are implicit and culturally variable. We propose a human-large language model (LLM) collaborative re-annotation framework for stabilizing multilingual speaker-attribute labels under practical resource constraints. Starting from a noisy corpus, we use LLMs to surface recurring annotation rationales through iterative interaction with experts, and apply disagreement-focused sampling for targeted re-annotation. Using this framework, we construct WhoSaidIt, a multilingual dataset covering nine speaker-attribute labels. We quantify divergence between original and revised annotations, benchmark recent LLMs, and analyze the effect of explicit rationales on model behavior. Our results reveal substantial cross-lingual differences in annotation decisions and demonstrate both the strengths and limitations of LLMs in speaker-attribute classification.
title WhoSaidIt: Human-LLM Collaborative Annotation for Text-Based Multilingual Speaker-Attribute Classification
topic Computation and Language
url https://arxiv.org/abs/2605.26070