Metraipolis › archive › dc_574bb512fe6b

Analyzing Occupational Gender Bias across the LLM Training Lifecycle — arXiv:2610.23587

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.23587
License (as archived)
CC0-1.0
Archived
2026-10-19T08:43:54.246Z
SHA-256 of the archived text
18467a1c39e77bf40f43d1dbdbae0dd65440ef0a206e2e69bc23d636cfc3fd2c
Stable URL of this page
https://metraipolis.com/documento/dc_574bb512fe6b

How to cite

Analyzing Occupational Gender Bias across the LLM Training Lifecycle — arXiv:2610.23587 — arXiv (en), https://arxiv.org/abs/2610.23587 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_574bb512fe6b (Metraipolis archive dc_574bb512fe6b, sha256 18467a1c39e77bf40f43d1dbdbae0dd65440ef0a206e2e69bc23d636cfc3fd2c)
BibTeX
@misc{dc574bb512fe6b,
  title = {Analyzing Occupational Gender Bias across the LLM Training Lifecycle — arXiv:2610.23587},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.23587},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_574bb512fe6b; sha256 18467a1c39e77bf40f43d1dbdbae0dd65440ef0a206e2e69bc23d636cfc3fd2c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.23587. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Post-training is widely assumed to reduce the social biases language models inherit from pretraining, yet direct evidence for this assumption remains mixed. In this paper, we examine how gender disparities evolve across the language modeling pipeline. In particular, we study occupational gender bias in OLMo 3, a fully open language model family, across its post-training stages by prompting each checkpoint to generate open-ended narratives about synthetic personas given their occupation and education. We consider two settings, a gender-given setting in which the profile states the persona's gender explicitly and a gender-assumed setting in which it must be inferred by the model. We find that when gender is stated, post-training increases semantic diversity for male-coded profiles, whereas for female-coded profiles the increase is smaller after SFT and not significant after DPO and RLVR. When gender is assumed by the model, post-training affects narratives about female and male personas similarly. The model's own gender assumptions, however, show a consistent female-preferential skew relative to United States labor-force demographics, except for heavily male-dominated occupations. Every successive post-training stage monotonically increases the odds of attributing a female label over the base model. A provenance analysis of the supervised fine-tuning corpus finds only a modest 7.7% female-token surplus, too small to explain the size of these downstream effects. The project code is available under: 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/2610.23587 — 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 2026-10-19T08:39:24.877Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).