Weekly Reads

FuseReg: Regularizing Layer Fusion Mitigates the Reconstruction-Generation Gap in Representation Autoencoders — Hongyang Du, Yunfei Xie et al.
Representation autoencoders (RAEs) reuse features from a pretrained visual encoder as reconstruction and diffusion latents, integrating strong visu...
Learning to Stop without Learning to Stop: Self-Supervised Confidence Training Improves Reasoning Efficiency — Parsa Hosseini, Akasha Tigalappanavara et al.
Reasoning models often generate very long reasoning traces, making inference computationally expensive. Existing approaches typically improve effic...
Gap-free Differentially Private PCA for Gaussian Data — Alina Ene, Huy L. Nguyen
We give a gap-free differentially private algorithm for the principal component analysis (PCA) problem with Gaussian data.
First-Order Stationarity of Reverse Diffusions — Zhifeng Chen, Chenyang Jiang, Yazhen Wang
Recent literature has shown a strong connection between optimization and sampling. We develop the corresponding first-order theory for diffusion mo...
Statistical attribute alignment for black-box generative AI via output post-processing — Kevin Jiang, Morgane Austern et al.
Generative AI systems are increasingly used, but aligning their outputs with user requirements poses a continuing challenge. Here, we aim to ensure...