无需人工标注，自动评估生成图像质量与提示匹配度。
ELIQ: A Label-Free Framework for Quality Assessment of Evolving AI-Generated Images
- 构建正负样本对，捕捉传统失真与AI生成特有失真模式。
- 在多基准测试中超越现有无标签方法，支持跨场景泛化。
- 适合需要大规模、持续评估生成图像质量的研究与应用。
生成式文本到图像模型发展迅猛，不断突破感知质量上限，导致以往收集的标签对新生成内容不再可靠。为此，我们提出ELIQ——一种面向动态演化生成图像的质量评估无标签框架。ELIQ聚焦视觉质量和提示-图像一致性，自动构建正样本及特定维度的负样本对，覆盖传统失真与AIGC特有的失真类型，实现无需人工标注的可迁移监督。基于这些样本对，通过指令微调将预训练多模态模型转化为质量感知判别器，并采用轻量级门控融合与质量查询变换器预测二维质量。在多个基准上的实验表明，ELIQ持续优于现有无标签方法，无需修改即可从AIGC泛化至UGC场景，为持续演进生成模型下的可扩展、无标签质量评估开辟了道路。代码将于发表后公开。
原文摘要 · Abstract (English)
Generative text-to-image models are advancing at an unprecedented pace, continuously shifting the perceptual quality ceiling and rendering previously collected labels unreliable for newer generations. To address this, we present ELIQ, a Label-free Framework for Quality Assessment of Evolving AI-generated Images. Specifically, ELIQ focuses on visual quality and prompt-image alignment, automatically constructs positive and aspect-specific negative pairs to cover both conventional distortions and AIGC-specific distortion modes, enabling transferable supervision without human annotations. Building on these pairs, ELIQ adapts a pre-trained multimodal model into a quality-aware critic via instruction tuning and predicts two-dimensional quality using lightweight gated fusion and a Quality Query Transformer. Experiments across multiple benchmarks demonstrate that ELIQ consistently outperforms existing label-free methods, generalizes from AI-generated content (AIGC) to user-generated content (UGC) scenarios without modification, and paves the way for scalable and label-free quality assessment under continuously evolving generative models. The code will be released upon publication.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。