3 papers
cs.LG2026
Continuous First, Discrete Later: VQ-VAEs Without Dimensional Collapse
Xinyu Zhao, Nikita Karagodin, Hamed Hassani +3
While many approaches to improve VQ-VAE performance focus on codebook size and utilization, the effect of dimensional collapse, where trained VQ-VAE representations live in an extr…
cs.IT2025
Approaching Rate-Distortion Limits in Neural Compression with Lattice Transform Coding
Eric Lei, Hamed Hassani, Shirin Saeedi Bidokhti
Neural compression has brought tremendous progress in designing lossy compressors with good rate-distortion (RD) performance at low complexity. Thus far, neural compression design…
cs.IT2025
Optimal Neural Compressors for the Rate-Distortion-Perception Tradeoff
Eric Lei, Hamed Hassani, Shirin Saeedi Bidokhti
Recent efforts in neural compression have focused on the rate-distortion-perception (RDP) tradeoff, where the perception constraint ensures the source and reconstruction distributi…