arXiv Computer Vision
Bridging Adversarial and Collaborative Learning for AI-Generated Image Quality Assessment
Read the original on arXiv Computer Vision →The Flow has not summarised this story yet — read it at arXiv Computer Vision.
Related stories
arXiv Computer Vision
Aug 27
PreResQ-R1: Response-Preference Disentangled Ranking-and-Scoring Reinforcement Optimization for Robust Visual Quality Assessment
PreResQ‑R1 introduces a Preference‑Response Disentangled Reinforcement Learning framework for Visual Quality Assessment that jointly optimizes absolute score regression and relative ranking consistency. It employs a dual‑branch reward system—modeling intra‑sample response coherence and inter‑sample preference alignment—trained with Group Relative Policy Optimization. The method extends to video quality assessment via a global‑temporal and local‑spatial data flow strategy, achieving state‑of‑the‑art results on 10 IQA and 5 VQA benchmarks with only 6K images and 28K videos, and provides human‑aligned reasoning traces.
arXiv Computer Vision
Sep 22
An Evolutionary Agentic Approach for Open-ended Image Quality Perception
arXiv:2609.22942v1 Announce Type: new Abstract: Generative models are rapidly expanding image quality assessment (IQA) beyond traditional fidelity factors to emerging dimensions such as physical plau...
arXiv Computer Vision
Sep 23
FoMo: Forking Moment in Generative Trajectory as a Perceptual Distance
arXiv:2609.25716v1 Announce Type: new Abstract: Reference-based image quality assessment (IQA) metrics aim to reflect how humans perceive the perceptual distance between a pair of images. To learn ho...
arXiv AI
Jun 2
Improving Visual Representation Alignment Generation with GRPO
arXiv:2606. 00583v1 Announce Type: cross Abstract: Recent diffusion transformers have demonstrated strong image synthesis capabilities but remain inefficient to train due to weak alignment between generative and discriminative representations.
arXiv AI
Sep 30
Mutually Adversarial Self-Training with Evolving Data for Unified Multimodal Models
The paper introduces MATE, a reinforcement‑learning‑based post‑training framework for unified multimodal models that lets the generation and understanding branches challenge each other instead of cooperating. In MATE, each branch proposes candidate outputs that the other must reproduce, and the solver is trained on the worst‑handled candidate, creating an evolving adversarial loop without a separate adversary. Experiments on Janus‑Pro‑1B show that MATE improves generation and understanding metrics, including GenEval (+2.4), DPG‑Bench (+1.7), and an average of nine understanding benchmarks (+0.7), while enhancing consistency across image‑text cycles.
arXiv AI
Jul 20
Debiasing Text-to-Image Evaluation via Implicit Cultural Alignment Reward Modeling
arXiv:2607. 15740v1 Announce Type: cross Abstract: As Text-to-Image (T2I) systems rapidly advance, evaluating the cultural authenticity of synthesized content has become increasingly important for fair and trustworthy generative AI.