[ACL 2024 Findings & ICLR 2024 WS] An Evaluator VLM that is open-source, offers reproducible evaluation, and inexpensive to use. Specifically designed for fine-grained evaluation on customized score rubric, Prometheus-Vision is a good alternative for human evaluation and GPT-4V evaluation.
I have noticed that certain citations for the related work section are absent in the paper, so I leave the paper information that comes to my mind (without shame, including ours..).
LLMScore: Unveiling the Power of Large Language Models in Text-to-Image Synthesis Evaluation (NeurIPS 2023) https://arxiv.org/abs/2305.11116
SelfEval: Leveraging the discriminative nature of generative models for evaluation (arXiv 2023) https://arxiv.org/abs/2311.10708
Mutual Information Divergence: A Unified Metric for Multimodal Generative Models (NeurIPS 2022) http://arxiv.org/abs/2205.13445
I have noticed that certain citations for the related work section are absent in the paper, so I leave the paper information that comes to my mind (without shame, including ours..).