method · first seen 6 Oct 2026
Sinkhorn divergence
Also: Sinkhorn distance
An entropy-regularized optimal transport distance computed with the Sinkhorn iteration, used as a differentiable loss between distributions.
Mentions per week
PapersCodeDiscussion
Earliest sources
- WASD: Wasserstein-based Knowledge Distillation for Large Language ModelsPapers · 6 Oct 2026 · Byeonghu Na, Donghyeok Shin, Yeongmin Kim, Mina Kang, Il-Chul Moon
- Unbalanced W-Flow: Unbalanced Wasserstein Drifting under Finite-Batch Marginal FluctuationsPapers · 13 Oct 2026 · Jongwook Jeon, Mingue Park, Jiwon Choi, Jaihoon Kim, Yunhong Min, Minhyuk Sung
- Second-order Sinkhorn Geometry and $\sqrt{n}$-InferencePapers · 27 Oct 2026 · Tristan Luca Saidi, Florian Gunsilius, Gonzalo Mena
- sDMD: Sinkhorn Distribution Matching DistillationPapers · 29 Oct 2026 · Jaihoon Kim, Lingxiao Li, Soshi Shimada, Roshni Cooper, Minhyuk Sung, Yiwei Zhao
Latest sources
- sDMD: Sinkhorn Distribution Matching DistillationPapers · 29 Oct 2026 · Jaihoon Kim, Lingxiao Li, Soshi Shimada, Roshni Cooper, Minhyuk Sung, Yiwei Zhao
- Second-order Sinkhorn Geometry and $\sqrt{n}$-InferencePapers · 27 Oct 2026 · Tristan Luca Saidi, Florian Gunsilius, Gonzalo Mena
- Unbalanced W-Flow: Unbalanced Wasserstein Drifting under Finite-Batch Marginal FluctuationsPapers · 13 Oct 2026 · Jongwook Jeon, Mingue Park, Jiwon Choi, Jaihoon Kim, Yunhong Min, Minhyuk Sung
- WASD: Wasserstein-based Knowledge Distillation for Large Language ModelsPapers · 6 Oct 2026 · Byeonghu Na, Donghyeok Shin, Yeongmin Kim, Mina Kang, Il-Chul Moon