← Back to all news
arXiv AI By Yingjia Wan, Lin Lin, Elisa Kreiss

How Humans and LLMs Read Gender into "Gender-Neutral" Physical Descriptions

Read the original on arXiv AI →

The study introduces GAPA, a dataset of 316 physical attributes with 14,706 gender-association ratings from 304 US annotators, showing that such descriptions carry structured gender associations. It evaluates 16 LLMs, finding they partially mirror human ratings but exhibit biases such as compressed distributions, weaker alignment for men, and asymmetric abstention toward non‑binary identities. A proxy model trained on these data is released and applied to analyze character descriptions in LitBank, illustrating the persistence of gendered interpretations in ostensibly neutral language.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

Related stories

arXiv Computation and Language
Sep 11

Alignment Reduces Expressed but Not Encoded Gender Bias: A Unified Framework and Study

The paper introduces a unified framework that simultaneously measures intrinsic (encoded) and extrinsic (expressed) gender bias in large language models using identical neutral prompts. It finds a consistent link between latent gender information and output bias, but shows that alignment via supervised fine‑tuning reduces expressed bias while leaving internal gender associations largely intact and reactivatable by adversarial prompts. The study also demonstrates that debiasing gains on structured benchmarks may not transfer to realistic tasks such as story generation.

By Nour Bouchouchi, Thibault Laugel, Xavier Renard, Christophe Marsala, Marie-Jeanne Lesot, Marcin Detyniecki
llmsfine-tuningbenchmarkssafety
More like this →
arXiv Computation and Language
Sep 17

Which Demographics do LLMs Default to During Annotation?

The paper investigates which demographic attributes large language models (LLMs) default to when annotating text without explicit demographic cues. By comparing non‑demographic, placebo‑conditioned, and demographic‑conditioned prompts on politeness and offensiveness tasks in the POPQUORN dataset, the authors find that LLMs exhibit notable gender, race, and age influences in their annotations. This contrasts with earlier studies that reported no such effects, highlighting the importance of considering demographic bias in LLM‑based annotation workflows.

By Johannes Sch\"afer, Aidan Combs, Christopher Bagdon, Jiahui Li, Nadine Probol, Lynn Greschner, Sean Papay, Yarik Menchaca Resendiz, Aswathy Velutharambath, Amelie W\"uhrl, Sabine Weber, Roman Klinger
llms
More like this →