Metraipolis › archive › dc_c07da3937a85
TruthLens: Object Hallucination Detection via Self-Evaluating Truthfulness Scores in LVLMs — arXiv:2608.05616
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.05616
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T10:36:01.644Z
- SHA-256 of the archived text
50ef8deb1a3585005d50962bad1bac34b140738a127d212acfc2cb006ab61179- Stable URL of this page
https://metraipolis.com/documento/dc_c07da3937a85
How to cite
TruthLens: Object Hallucination Detection via Self-Evaluating Truthfulness Scores in LVLMs — arXiv:2608.05616 — arXiv (en), https://arxiv.org/abs/2608.05616 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c07da3937a85 (Metraipolis archive dc_c07da3937a85, sha256 50ef8deb1a3585005d50962bad1bac34b140738a127d212acfc2cb006ab61179)
BibTeX
@misc{dcc07da3937a85,
title = {TruthLens: Object Hallucination Detection via Self-Evaluating Truthfulness Scores in LVLMs — arXiv:2608.05616},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.05616},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c07da3937a85; sha256 50ef8deb1a3585005d50962bad1bac34b140738a127d212acfc2cb006ab61179},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.05616. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Despite the remarkable progress of large vision language models (LVLMs), object hallucination remains a fundamental challenge that hinders their trustworthy deployment. A key finding motivates our work: real and hallucinated object tokens are clearly separable in hidden representations, yet this separability is largely lost at the language-modeling (LM) head. We propose TruthLens, a self-evaluation framework that teaches the LM head to expose a per-object truthfulness signal without any auxiliary model or additional inference cost. Concretely, a rarely-used special token is repurposed as a reference token. For each object-token position, we extract the log-probability assigned to this special token by the LM head, and define its difference from a predefined constant as the truthfulness score. The model is then fine-tuned with an MSE objective that drives scores toward 1 for real objects and 0 for hallucinated ones, while a divergence constraint preserves the original generation capability. Despite being trained on only a limited set of object categories, TruthLens generalizes effectively to benchmarks with substantially larger label spaces. Extensive experiments across multiple LVLMs demonstrate state-of-the-art performance; notably, on Qwen2.5-VL-7B, TruthLens outperforms the previous best method on MS-COCO by over 17\% in AUROC. Our code is available at this https URL . Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-07 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2608.05616 — 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 2026-10-31T10:23:30.987Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-08-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).