Yu Shi

When Latents Forget Pixels: Restoring Fidelity in Diffusion Transformer Super-Resolution

This work proposes a pixel-grounded super-resolution (PGSR) framework that preserves LR-observed pixel evidence before VAE compression and reuses it throughout restoration and improves the realism--fidelity trade-off and produces more faithful, visually convincing results than existing latent generative SR approaches.