Metraipolis › archive › dc_4091c8883fc4

Conformal Fair Decoding for Large Language Model Recommenders — arXiv:2611.10285

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.10285
License (as archived): CC0-1.0
Archived: 2026-11-10T18:40:52.643Z
SHA-256 of the archived text: 47ff5494c3cc26340ba572fa80523af64bb5dbdc0bb1212b92877d2a5ae72f80
Stable URL of this page: https://metraipolis.com/documento/dc_4091c8883fc4

How to cite

Conformal Fair Decoding for Large Language Model Recommenders — arXiv:2611.10285 — arXiv (en), https://arxiv.org/abs/2611.10285 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4091c8883fc4 (Metraipolis archive dc_4091c8883fc4, sha256 47ff5494c3cc26340ba572fa80523af64bb5dbdc0bb1212b92877d2a5ae72f80)

BibTeX

@misc{dc4091c8883fc4,
  title = {Conformal Fair Decoding for Large Language Model Recommenders — arXiv:2611.10285},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.10285},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4091c8883fc4; sha256 47ff5494c3cc26340ba572fa80523af64bb5dbdc0bb1212b92877d2a5ae72f80},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.10285. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language models (LLMs) can generate personalized recommendations without task-specific training, but changing a protected attribute while holding interaction history fixed can alter the semantic content of recommended items. We study fairness control when model weights and full-vocabulary logits are unavailable. We propose Fair- Rec, an item-level decoding framework that uses an offline-calibrated semantic-disparity threshold and controls candidate selection as recommendation lists are generated. A hard variant filters candidates against the threshold; a soft variant trades off candidate likelihood and disparity while adapting a penalty using observed violations. Under an explicit uniform enforceability assumption, we establish a pathwise O(1/T) upper bound on the excess empirical violation rate of the soft decoder. Experiments on MovieLens-1M and Amazon Movies & TV using Qwen3-8B and Llama3.1-8B-Instruct show that the soft variant lowers measured counterfactual disparity relative to zero- =shot recommendation, UP5, and FACTER, while retaining competitive HR@10 and NDCG@10.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2611.10285 — 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 2026-11-10T18:37:37.485Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).