Metraipolis › archive › dc_bc4beaaf4bad

Mitigating Object Hallucination in Large Vision-Language Models via False Discovery Controlled Visual Data Splitting — arXiv:2609.38979

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.38979
License (as archived)
CC0-1.0
Archived
2026-10-07T14:45:40.972Z
SHA-256 of the archived text
dacad25b6f124cc0d374435ff6e775cae29a4027d946610c0e668e346436fda0
Stable URL of this page
https://metraipolis.com/documento/dc_bc4beaaf4bad

How to cite

Mitigating Object Hallucination in Large Vision-Language Models via False Discovery Controlled Visual Data Splitting — arXiv:2609.38979 — arXiv (en), https://arxiv.org/abs/2609.38979 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bc4beaaf4bad (Metraipolis archive dc_bc4beaaf4bad, sha256 dacad25b6f124cc0d374435ff6e775cae29a4027d946610c0e668e346436fda0)
BibTeX
@misc{dcbc4beaaf4bad,
  title = {Mitigating Object Hallucination in Large Vision-Language Models via False Discovery Controlled Visual Data Splitting — arXiv:2609.38979},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.38979},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bc4beaaf4bad; sha256 dacad25b6f124cc0d374435ff6e775cae29a4027d946610c0e668e346436fda0},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.38979. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Multiple object hallucination, where large vision-language models (LVLMs) generate objects not supported by the visual input, is a persistent challenge caused by visual uncertainty during decoding. Existing methods reduce hallucinations using contrastive signals, but they rely on heuristics and lack principled control of false positives at the image level. To address this, we propose False Discovery Rate-COntRol of HALlucination (CORAL), a training-free framework that models visual uncertainty using an uncertainty-aware visual data splitting strategy and leverages mirror statistics to quantify visual contrast during decoding. By computing mirror statistics from paired, symmetrically perturbed visual inputs, CORAL estimates spurious object predictions and sets a data-driven threshold to control the expected fraction of false discoveries per image, suppressing hallucinations while retaining high power for truly grounded objects. The framework is flexible, supports multiple LVLMs, and mitigates hallucinations without retraining or supervision. Extensive experiments on multiple benchmarks with several evaluation metrics demonstrate that CORAL consistently outperforms state-of-the-art methods, providing more reliable and robust hallucination control. Code is available at: this https URL

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-01 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.38979 — 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 2026-10-07T14:19:00.643Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).