Metraipolis › archive › dc_f5601227151f

Reliability-Aware Sexism Detection: Combining DPO with Annotator Agreement and Token-Level Confidence Scoring — arXiv:2608.12330

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.12330
License (as archived)
CC0-1.0
Archived
2026-10-19T04:24:21.408Z
SHA-256 of the archived text
91e636dd5bb280bbfd8e03653fc5e1c71985276a95b27c6b8837ad1abf7b35f4
Stable URL of this page
https://metraipolis.com/documento/dc_f5601227151f

How to cite

Reliability-Aware Sexism Detection: Combining DPO with Annotator Agreement and Token-Level Confidence Scoring — arXiv:2608.12330 — arXiv (en), https://arxiv.org/abs/2608.12330 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f5601227151f (Metraipolis archive dc_f5601227151f, sha256 91e636dd5bb280bbfd8e03653fc5e1c71985276a95b27c6b8837ad1abf7b35f4)
BibTeX
@misc{dcf5601227151f,
  title = {Reliability-Aware Sexism Detection: Combining DPO with Annotator Agreement and Token-Level Confidence Scoring — arXiv:2608.12330},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.12330},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f5601227151f; sha256 91e636dd5bb280bbfd8e03653fc5e1c71985276a95b27c6b8837ad1abf7b35f4},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.12330. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

The detection of online sexism remains an open problem. Sexism detection is inherently subjective, yet most existing systems reduce multi-annotator labels to a single majority decision and treat all instances uniformly. This ignores two informative signals: annotator agreement and model uncertainty. We propose RA-DPO (Reliability-Aware Direct Preference Optimization), which integrates annotator agreement, model confidence, and a token-level uncertainty signal into a single reliability score. RA-DPO uses this score to select high-value preference pairs during training and to support inference-time abstention, which allows the model to trade coverage for accuracy. We evaluate RA-DPO on 6,920 multilingual posts from EXIST 2023, fine-tune OpenAI gpt-4o base via DPO, and validate on two open-weight 3B models (Llama, Qwen). Results show that training on the top 30% most reliable pairs matches full-data DPO, which indicates that reliability-aware selection can reduce training cost without sacrificing performance. At inference, selective prediction reaches 96.2% accuracy at 50% coverage in the true-agreement setting and 88.7% in the deployable predicted-agreement setting, both exceeding the 85.3% no-agreement baseline. These results suggest that accounting for annotation uncertainty is beneficial for both efficient training and reliable deployment in subjective classification.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-14 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2608.12330 — 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 2026-10-19T04:17:42.361Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-14?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).