PACE (Perceptual Agentic Collaborative Evolution), a training-free multi-agent framework that formulates open-ended IQA as explicit protocol construction, is proposed and consistently improves its MLLM backbone, achieving competitive performance across diverse IQA settings, and reduces the Holistic Override Rate.
Abstract
Generative models are rapidly expanding image quality assessment (IQA) beyond traditional fidelity factors to emerging dimensions such as physical plausibility and text-rendering correctness. However, existing IQA models rely on fixed definitions and heavy supervision, making them difficult to extend to open-ended perceptual dimensions. We identify holistic bias as an important limitation: when scoring an unseen dimension, models reuse generic quality priors, leading to scoring errors and rank inversion. To address this, we propose PACE (Perceptual Agentic Collaborative Evolution), a training-free multi-agent framework that formulates open-ended IQA as explicit protocol construction. Given a target dimension, PACE uses collaborative agents to construct an evaluation protocol composed of verifiable Visual Question Answering (VQA) probes, grounding evaluation in concrete visual evidence rather than holistic impressions. The resulting protocol is calibrated using only four human-annotated images per dimension, while a dual-track scoring mechanism aligns model perception with human scoring scales. Across traditional IQA, structural fidelity, context-aware aesthetics, and newly defined open-ended dimensions, PACE consistently improves its MLLM backbone, achieving competitive performance across diverse IQA settings, and reduces the Holistic Override Rate (HOR) from 44.4\% to 8.6\%.
A multi-dimensional image quality assessment framework for T2I tasks that can not only distinguish overall performance differences across models, but also provide interpretable results at the level of individual dimensions is proposed.
GS-IQA is proposed, a framework that reformulates IQA as a progressive Where--What--How diagnosis, emulating the human perceptual process from an initial glance to closer scrutiny, and consistently surpasses state-of-the-art methods in distortion localization, recognition, and severity estimation.
Ao-Ting Zhang, Ming-Ze Gao, Dong-Bao Yang et al.· 0 citations
VIEScore2, a unified evaluator for image generation and editing tasks with optional conditioning images, outperforms both general-purpose VLMs and specialized spatial evaluators on three of six benchmarks in per-image grid IoU and ranks among the top three on five, including datasets beyond its training sources.
Xian-Da Du, Max W.F. Ku, Weiming Ren et al.· 0 citations
AI-generated image quality assessment (AIGIQA) requires jointly reasoning about perceptual fidelity and prompt alignment, two quality dimensions that are often treated as independent in existing AIGIQA models. However, by re-examining human ratings, we uncover a previously overlooked phenomenon: the two dimensions are...
The proposed AEKI framework translates formal art principles into structured, machine-executable instructions and indicates that domain knowledge can be incorporated into LVLM evaluation through a transparent rule-based layer, supporting applications in digital curation, computational aesthetics, and art education.
GenPuzzle provides a testbed for measuring progress from image rendering toward visual problem solving, and uses task-specific evaluation protocols that select the automatic judge by measuring agreement with human reference scores.
Known for his clear and elegant writing style, Bertsekas shaped fields from control and optimization to large-scale computation and artificial intelligence.