Metraipolis › archive › dc_982a589ee950
BEiTScore++: Reference-free Image Captioning Evaluation with BEiT-3 and Conformalized Quantile Regression — arXiv:2611.09505
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.09505
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-09T22:35:06.779Z
- SHA-256 of the archived text
13b41804a3f890cec2a916e62919e50e259b915cf6e79f84d6c24b6498a5540b- Stable URL of this page
https://metraipolis.com/documento/dc_982a589ee950
How to cite
BEiTScore++: Reference-free Image Captioning Evaluation with BEiT-3 and Conformalized Quantile Regression — arXiv:2611.09505 — arXiv (en), https://arxiv.org/abs/2611.09505 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_982a589ee950 (Metraipolis archive dc_982a589ee950, sha256 13b41804a3f890cec2a916e62919e50e259b915cf6e79f84d6c24b6498a5540b)
BibTeX
@misc{dc982a589ee950,
title = {BEiTScore++: Reference-free Image Captioning Evaluation with BEiT-3 and Conformalized Quantile Regression — arXiv:2611.09505},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.09505},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_982a589ee950; sha256 13b41804a3f890cec2a916e62919e50e259b915cf6e79f84d6c24b6498a5540b},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.09505. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Reliable uncertainty estimation is important for deploying evaluation metrics in real-world settings, where both input ambiguity and domain shifts are common. In this work, we propose an uncertainty-aware extension of the BEiTScore image captioning metric, producing conformally calibrated confidence intervals alongside alignment scores. Our approach models predictive distributions over human judgments, enabling both accurate scoring and principled uncertainty quantification in a single forward pass. We evaluate our BEiTScore++ metric across multiple benchmarks with human-annotated ratings, showing that it preserves the ranking correlation of BEiTScore, while producing intervals whose width tracks the prediction error more closely than Monte Carlo dropout on most benchmarks. The estimated uncertainty also grows under image degradation and on out-of-domain artwork images, making BEiTScore++ a practical tool for uncertainty-aware evaluation of image-caption alignment. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2611.09505 — 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 2026-11-09T22:30:47.335Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).