AI News HubLIVE
Original source2 min read

BEiTScore: Reference-free Image Captioning Evaluation with an Efficient Cross-Encoder Model

BEiTScore is a new reference-free image captioning evaluation metric based on a lightweight cross-encoder initialized from a VQA model checkpoint, using adversarial LLM-based data augmentation to achieve state-of-the-art performance with high efficiency.

SourcearXiv Computer VisionAuthor: Gon\c{c}alo Gomes, Bruno Martins, Chrysoula Zerva

[2605.21728] BEiTScore: Reference-free Image Captioning Evaluation with an Efficient Cross-Encoder Model

[Submitted on 20 May 2026]

Title:BEiTScore: Reference-free Image Captioning Evaluation with an Efficient Cross-Encoder Model

View a PDF of the paper titled BEiTScore: Reference-free Image Captioning Evaluation with an Efficient Cross-Encoder Model, by Gon\c{c}alo Gomes and 2 other authors

View PDF HTML (experimental)

Abstract:Image captioning evaluation remains a significant challenge, as vision-language models evolve toward more challenging capabilities such as generating long-form and context-rich descriptions. State-of-the-art evaluation metrics involve extensive computational costs associated with the use of Large Language Models (LLMs) as judges, or instead suffer from the limitations of standard CLIP-based encoders, such as strict token limits, lack of fine-grained sensitivity, or lack of compositional generalization by treating captions as ``bags-of-words.'' We propose a new learned metric that tackles the aforementioned challenges, based on a lightweight cross-encoder that is initialized from a visual question-answering model checkpoint, balancing a strong weight initialization with computational efficiency. Our training scheme uses a carefully assembled data mixture for supervised learning, featuring adversarial LLM-based data augmentations to enhance model sensitivity to fine-grained visual-linguistic errors. We also introduce a new benchmark designed to assess detailed captioning evaluation across diverse scenarios. Experimental results demonstrate that the proposed metric achieves state-of-the-art performance while maintaining the efficiency required for large-scale benchmarking, quality-aware decoding, or reward guidance.

Subjects:

Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL); Machine Learning (cs.LG)

Cite as: arXiv:2605.21728 [cs.CV]

(or arXiv:2605.21728v1 [cs.CV] for this version)

https://doi.org/10.48550/arXiv.2605.21728

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Gonçalo Emanuel Cavaco Gomes [view email] [v1] Wed, 20 May 2026 20:43:38 UTC (1,113 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled BEiTScore: Reference-free Image Captioning Evaluation with an Efficient Cross-Encoder Model, by Gon\c{c}alo Gomes and 2 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.CV

new | recent | 2026-05

Change to browse by:

cs cs.CL cs.LG

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Loading...

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)