Cross-Domain Lossy Compression via Rate- and Classification-Constrained Optimal Transport

ICLROral2026

Authors
Nam Nguyen, Thinh Nguyen, Bella Bose
Affiliation
Oregon State University
Venue
ICLR 2026
Track
Oral

TL;DR

We study cross-domain lossy compression via constrained optimal transport with rate and classification constraints, derive closed-form tradeoffs, extend to perception divergences, and validate with deep restoration and inpainting experiments.

Opening excerpt from the authors’ abstract. source

Read the paper

Topics

optimal transport

← All ICLR 2026 Oral papers · Browse the whole archive