REVIEW 4 cited by
EVALALIGN: Supervised Fine-Tuning Multimodal LLMs with Human-Aligned Data for Evaluating Text-to-Image Models
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
The recent advancements in text-to-image generative models have been remarkable. Yet, the field suffers from a lack of evaluation metrics that accurately reflect the performance of these models, particularly lacking fine-grained metrics that can guide the optimization of the models. In this paper, we propose EvalAlign, a metric characterized by its accuracy, stability, and fine granularity. Our approach leverages the capabilities of Multimodal Large Language Models (MLLMs) pre-trained on extensive data. We develop evaluation protocols that focus on two key dimensions: image faithfulness and text-image alignment. Each protocol comprises a set of detailed, fine-grained instructions linked to specific scoring options, enabling precise manual scoring of the generated images. We supervised fine-tune (SFT) the MLLM to align with human evaluative judgments, resulting in a robust evaluation model. Our evaluation across 24 text-to-image generation models demonstrate that EvalAlign not only provides superior metric stability but also aligns more closely with human preferences than existing metrics, confirming its effectiveness and utility in model assessment.
Forward citations
Cited by 4 Pith papers
-
Harm is not Universal: Community-Specific Toxicity Detection is Urgently Needed
SoTA T2I toxicity detectors miss ~35% of disability-community harms; zero-shot CTD fails below random, while ICL/VQA/LoRA improve but stay well below general TD performance.
-
Debiasing Text-to-Image Evaluation via Implicit Cultural Alignment Reward Modeling
A 4.2B-parameter reward model with skip-connection cross-attention to early visual tokens achieves 80.54% pairwise accuracy on CulturalFrames, beating GPT-4o and VQAScore at 10x lower inference cost.
-
Multimodal LLMs as Customized Reward Models for Text-to-Image Generation
LLaVA-Reward extracts reward scores from the hidden states of a multimodal LLM with a skip-connection cross-attention head, and reports state-of-the-art text-to-image evaluation across alignment, fidelity, and safety.
-
DIMCIM: A Quantitative Evaluation Framework for Default-mode Diversity and Generalization in Text-to-Image Generative Models
A new evaluation framework, DIMCIM, measures default-mode diversity and prompted generalization in text-to-image models, finding a scale trade-off and a 0.85 correlation between default diversity and training data diversity.
Discussion (0). Continue with ORCID to comment.