Signals 4 · free daily AI digest

FuseReg: Regularizing Layer Fusion Mitigates the Reconstruction-Generation Gap in Representation Autoencoders

Paper recorded by Signals 4 on 2026-09-25 in cs.CV. Abstract reproduced from arXiv; link to the original below.

Published 2026-09-25 on arXiv · recorded by Signals 4 on 2026-09-28

Category: cs.CV · 计算机视觉 · first seen 2026-09-28

Abstract

Representation autoencoders (RAEs) reuse features from a pretrained visual encoder as reconstruction and diffusion latents, integrating strong visual representations into image generation. However, RAEs still need to decide which encoder layers form the shared latent space for the generator and pixel decoder. This choice involves a trade-off. Shallower layers tend to preserve fine pixel details be

Read on arXiv →

#17 most recent of 342 cs.CV papers we have recorded · ↑ newer: Remote Sensing Sparse-View 3D Gaussian Splatting via Depth Image-Based · ↓ older: GraphWrit3R: End-to-End 3D Scene Graph Writing
Cite this page: FuseReg: Regularizing Layer Fusion Mitigates the Reconstruction-Generation Gap in Representation Autoencoders: the #17 most recent of 342 cs.CV papers we have recorded (as of 2026-09-25). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/fusereg-regularizing-layer-fusion-mitigates-the-reconstruction-generation-gap-in.html
Free to quote with attribution to “Signals 4 (Signals API)”. Machine-readable: papers.json
Related: More cs.CV papers · arXiv signals · All papers · Today in AI
Get 4 AI signals a day by email — free.
Subscribe free → See all plans →
Get 4 AI signals a day by email — free
All models · All repos · By company · Daily editions