AI RESEARCH
AvAtar: Learning to Align via Active Optimal Transport
arXiv CS.LG
•
ArXi:2605.24395v1 Announce Type: new Alignment plays a fundamental role in many machine learning problems, such as multi-network analysis, multimodal learning, and point cloud registration. Recent works increasingly leverage optimal transport (OT) for distributional alignment, whose effectiveness largely depends on sparse supervision that is hard or costly to obtain in practice. Existing works, however, largely overlook how to actively acquire high-quality supervision to improve their alignment performance under OT frameworks.