--- title: "Semantically Aware Optimal Transport for Dense Label Transfer" description: "The authors propose SAOT, which learns the transport cost itself through bilevel training with sparse target point labels. 1 reference." authors: ["Preeti", "Kiran Ravish", "Ankita Kushwaha", "Pawan Kumar"] year: 2026 venue: "Proceedings of the 4th Workshop on Advances in Language and Vision Research (ALVR)" source: "https://aclanthology.org/2026.alvr-main.3/" doi: "10.18653/v1/2026.alvr-main.3" licence: "CC BY 4.0 (ACL Anthology)" licence_url: "https://creativecommons.org/licenses/by/4.0/" summary_licence: "Elamur Cosmic License" canonical: "https://research.elamur.ai/free/en/papers/preeti-2026-semantically-aware-optimal-transport-dense/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/preeti-2026-semantically-aware-optimal-transport-dense/) [home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/42/) / Semantically Aware Optimal Transport for Dense Label Transfer # Semantically Aware Optimal Transport for Dense Label Transfer **Authors:** Preeti, Kiran Ravish, Ankita Kushwaha and Pawan Kumar · **Year:** 2026 · **Venue:** [Proceedings of the 4th Workshop on Advances in Language and Vision Research (ALVR)](https://md.research.elamur.ai/free/en/venues/alvr/) ([ALVR 2026](https://md.research.elamur.ai/free/en/venues/alvr/2026/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/) **Links:** [Publisher's page](https://aclanthology.org/2026.alvr-main.3/) · [PDF](https://aclanthology.org/2026.alvr-main.3.pdf) · [DOI 10.18653/v1/2026.alvr-main.3](https://doi.org/10.18653/v1/2026.alvr-main.3) · [OpenAlex](https://openalex.org/W7166812361) **Citations:** 0 citations (OpenAlex, 2026-10-14) ## Summary The paper addresses dense label transfer for semantic segmentation: projecting per-pixel class labels from a labeled source domain onto an unlabeled target domain while the vision transformer backbone stays frozen. Standard optimal-transport (OT) baselines match patches with a fixed similarity measure. Under domain shift, this confuses lookalike but semantically different regions, ignores class-frequency differences and scales poorly to thousands of tokens. The authors propose SAOT, which learns the transport cost itself through bilevel training with sparse target point labels. The cost mixes a trainable appearance metric with a penalty based on source class prototypes. It is plugged into an entropic, unbalanced OT problem solved over k-nearest-neighbour blocks. The resulting pseudo-labels train a residual MLP decoder, which is then refined by EMA-teacher self-training. On GTA5→Cityscapes with frozen DINOv2 ViT-L/14, the full pipeline reaches 25.7% mIoU, about 3.8× nearest-neighbour transfer. The authors report this as the best result among frozen-backbone methods. This shows that learning the matching cost, not the transport machinery alone, drives the gains. ## Key points - **Setup:** GTA5→Cityscapes at 504×1008 resolution (2,592 tokens per image) with frozen DINOv2 ViT-L/14. The block-sparse solver uses k = 128 neighbours, cutting coupling size from O(nm) to O(nk). - **Main results:** Fixed-cost OT, UOT and RWOT score only 6.5–6.8% mIoU, while balanced OT collapses to 0%. Raw SAOT reaches 12.6%, rising to 15.8% when aggregating 3 source images. SAOT+Decoder reaches 25.7% and labels 500 images in 1.2 s. - **Per-class results:** Road scores 90.3%, car 76.2% and building 71.5%. Six small or rare classes stay at 0% IoU, including traffic sign, rider and bicycle. - **Refinement and transport behaviour:** EMA refinement with hard labels and label smoothing lifts mIoU from 23.7% to 25.7%, whereas soft KL targets drop it to 9.9%. The learned cost cuts mass sent to wrong classes from 68% to 41%. - **Second benchmark and context:** On PASCAL VOC 2012 train→val with ViT-B/16, the full pipeline scores 47.5% versus KNN-Global at 45.6%. Fully fine-tuned UDA methods (DAFormer, HRDA, MIC) reach 68–76% and Rein reaches 78.4%. SAOT uses a 5.3M-parameter head and trains in about 6 GPU-hours. *Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.* ## Institutions - [International Institute of Information Technology, Hyderabad](https://md.research.elamur.ai/free/en/universities/international-institute-of-information-technology-hyderabad/) — University, Hyderabad, India · 38 works · 1,821 citations 📚 Shelves (3): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [👁️ Computer Vision and Pattern Recognition](https://md.research.elamur.ai/free/en/shelves/computer-science/computer-vision-and-pattern-recognition/) · 2,869 works ## Cite Cite as: Preeti, Kiran Ravish, Ankita Kushwaha and Pawan Kumar. 2026. Semantically Aware Optimal Transport for Dense Label Transfer. Proceedings of the 4th Workshop on Advances in Language and Vision Research (ALVR). The paper itself, not this page. BibTeX from the library's record of the work. ```bibtex @inproceedings{preeti2026semantically, title = {{Semantically Aware Optimal Transport for Dense Label Transfer}}, author = {Preeti, Kiran Ravish, Ankita Kushwaha and Pawan Kumar}, booktitle = {Proceedings of the 4th Workshop on Advances in Language and Vision Research (ALVR)}, year = {2026}, doi = {10.18653/v1/2026.alvr-main.3}, url = {https://doi.org/10.18653/v1/2026.alvr-main.3} } ``` - [← previous Semantic, Syntactic, Lexical: What Makes QA Augmentation Work in Limited Quantity?](https://md.research.elamur.ai/free/en/papers/rachmat-2026-semantic-syntactic-lexical-what-makes/) - [next → SemBench: A Universal Semantic Framework for LLM Evaluation](https://md.research.elamur.ai/free/en/papers/zubillaga-2026-sembench-universal-semantic-framework-llm/) - [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/42/) ## References 1 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet. 1. Jason Altschuler, Jonathan Weed, and Philippe Rigollet. 2017. Near-linear time approximation algorithms for optimal transport via sinkhorn iteration. In Proceedings of the 31st International Conference on Neural Information Processing Systems, NIPS’17, page 1961–1971, Red Hook, NY, USA. Curran Assoc… [find](https://scholar.google.com/scholar?q=Jason+Altschuler%2C+Jonathan+Weed%2C+and+Philippe+Rigollet%2E+2017%2E+Near%2Dlinear+time+approximation+algorithms+for+optimal+transport+via+sinkhorn+iteration%2E+In+Proceedings+of+the+31st+International+Conferenc) *Updated 1 November 2026*