FinPercep-RM with co-evolutionary curriculum learning stabilizes RL training for real-world image super-resolution by supplying local degradation feedback and reducing reward hacking.
Component divide-and-conquer for real-world image super-resolution
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2verdicts
UNVERDICTED 2representative citing papers
Q-DeepSight proposes a think-with-image multimodal CoT framework trained via RL with perceptual curriculum rewards and evidence gradient filtering to achieve SOTA IQA performance and enable training-free perceptual refinement in image generation.
citing papers explorer
-
FinPercep-RM: A Fine-grained Reward Model and Co-evolutionary Curriculum for RL-based Real-world Super-Resolution
FinPercep-RM with co-evolutionary curriculum learning stabilizes RL training for real-world image super-resolution by supplying local degradation feedback and reducing reward hacking.
-
Q-DeepSight: Incentivizing Thinking with Images for Image Quality Assessment and Refinement
Q-DeepSight proposes a think-with-image multimodal CoT framework trained via RL with perceptual curriculum rewards and evidence gradient filtering to achieve SOTA IQA performance and enable training-free perceptual refinement in image generation.