@inproceedings{slobodkin-etal-2025-refvnli,
    title = "{R}ef{VNLI}: Towards Scalable Evaluation of Subject-driven Text-to-image Generation",
    author = "Slobodkin, Aviv  and
      Taitelbaum, Hagai  and
      Bitton, Yonatan  and
      Gordon, Brian  and
      Sokolik, Michal  and
      Bitton Guetta, Nitzan  and
      Gueta, Almog  and
      Rassin, Royi  and
      Lischinski, Dani  and
      Szpektor, Idan",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2025",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-luhme/2025.findings-emnlp.447/",
    doi = "10.18653/v1/2025.findings-emnlp.447",
    pages = "8420--8438",
    ISBN = "979-8-89176-335-7",
    abstract = "Subject-driven text-to-image (T2I) generation aims to produce images that align with a given textual description, while preserving the visual identity from a referenced subject image. Despite its broad downstream applicability{---}ranging from enhanced personalization in image generation to consistent character representation in video rendering{---}progress in this field is limited by the lack of reliable automatic evaluation. Existing methods either assess only one aspect of the task (i.e., textual alignment or subject preservation), misalign with human judgments, or rely on costly API-based evaluation. To address this gap, we introduce RefVNLI, a cost-effective metric that evaluates both textual alignment and subject preservation in a single run. Trained on a large-scale dataset derived from video-reasoning benchmarks and image perturbations, RefVNLI outperforms or statistically matches existing baselines across multiple benchmarks and subject categories (e.g., \textit{Animal}, \textit{Object}), achieving up to 6.4-point gains in textual alignment and 5.9-point gains in subject preservation."
}