Metraipolis › archive › dc_0752affb08eb

CheXPoT: Post-Hoc Visual Verification and Correction of VLM-Generated Radiology Reports — arXiv:2611.00785

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.00785
License (as archived): CC0-1.0
Archived: 2026-11-03T18:22:40.367Z
SHA-256 of the archived text: 8460848a5ceeefed960c5e1cc355785c4245e81b2535b837ae2e326ec60cf25d
Stable URL of this page: https://metraipolis.com/documento/dc_0752affb08eb

How to cite

CheXPoT: Post-Hoc Visual Verification and Correction of VLM-Generated Radiology Reports — arXiv:2611.00785 — arXiv (en), https://arxiv.org/abs/2611.00785 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0752affb08eb (Metraipolis archive dc_0752affb08eb, sha256 8460848a5ceeefed960c5e1cc355785c4245e81b2535b837ae2e326ec60cf25d)

BibTeX

@misc{dc0752affb08eb,
  title = {CheXPoT: Post-Hoc Visual Verification and Correction of VLM-Generated Radiology Reports — arXiv:2611.00785},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.00785},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0752affb08eb; sha256 8460848a5ceeefed960c5e1cc355785c4245e81b2535b837ae2e326ec60cf25d},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.00785. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Vision-language models (VLMs) have shown promise for chest radiograph report generation, but adoption remains limited because of hallucinated findings that are not grounded in the image. Correcting these hallucinations and improving overall report quality remains a significant challenge. Prior work has addressed this problem by refining models with reinforcement learning, preference optimization, or chain-of-thought supervision, often at substantial compute cost. To address this gap, we introduce CheXPoT, a program-of-thought (PoT) framework that verifies each generated finding against image features and removes those it cannot ground. CheXPoT decomposes generated reports into predicates with RadGraph and an LLM, grounds these predicates against a DenseNet-121 and Grad-CAM visual evidence table, executes PoT verification code, and rewrites reports after removing unsupported predicates. On a 4,145-image held-out split, CheXPoT yields statistically significant precision gains on 8 of 14 CheXpert classes and F1 gains on 5, with no significant decreases in either metric. CheXPoT lifts micro-average precision from 19.2% (MedGemma-4B) to 46.8% (MedGemma-4B + CheXPoT) and micro-average F1 from 23.0% to 49.4%, absolute gains of 27.6% and 26.4% respectively. We also tested CheXPoT on other VLM report generators. On LLaVA-Med, CheXPoT lifts micro-average precision from 4.5% to 17.7% and micro-average F1 from 6.3% to 19.4%. On MAIRA-2, CheXPoT lifts micro-average precision from 15.6% to 42.1% and micro-average F1 from 21.0% to 45.3%. Our results suggest that inexpensive post-hoc visual verification may improve VLM trustworthiness by correcting unsupported findings from generated reports. CheXPoT lays a foundation for tools that help validate VLM-generated radiology reports and reduce unsupported findings. These innovations may subsequently help increase trust and adoption of VLM tools in Medicine.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL) ; Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2611.00785 — 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 2026-11-03T18:10:07.901Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).