{"paper":{"title":"U-Mamba: Enhancing Long-range Dependency for Biomedical Image Segmentation","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"U-Mamba pairs convolutional layers with state space models to capture long-range dependencies more effectively than prior CNN or Transformer networks for biomedical image segmentation.","cross_cats":["cs.CV","cs.LG"],"primary_cat":"eess.IV","authors_text":"Bo Wang, Feifei Li, Jun Ma","submitted_at":"2024-01-09T18:53:20Z","abstract_excerpt":"Convolutional Neural Networks (CNNs) and Transformers have been the most popular architectures for biomedical image segmentation, but both of them have limited ability to handle long-range dependencies because of inherent locality or computational complexity. To address this challenge, we introduce U-Mamba, a general-purpose network for biomedical image segmentation. Inspired by the State Space Sequence Models (SSMs), a new family of deep sequence models known for their strong capability in handling long sequences, we design a hybrid CNN-SSM block that integrates the local feature extraction p"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"The results reveal that U-Mamba outperforms state-of-the-art CNN-based and Transformer-based segmentation networks across all tasks.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the hybrid CNN-SSM block will reliably improve long-range dependency capture and generalization across diverse biomedical datasets without introducing training instability or requiring dataset-specific tuning beyond the claimed self-configuring mechanism.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"U-Mamba is a hybrid CNN-SSM architecture that outperforms prior CNN and Transformer networks on biomedical image segmentation tasks by efficiently modeling long-range dependencies.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"U-Mamba pairs convolutional layers with state space models to capture long-range dependencies more effectively than prior CNN or Transformer networks for biomedical image segmentation.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"eff7e42d009be3e3b54d2b3eb396e21d2365c90aa40600c1830ef488817065e3"},"source":{"id":"2401.04722","kind":"arxiv","version":1},"verdict":{"id":"13dbf27d-86be-4b45-b37f-262e98421dd8","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T12:32:41.682765Z","strongest_claim":"The results reveal that U-Mamba outperforms state-of-the-art CNN-based and Transformer-based segmentation networks across all tasks.","one_line_summary":"U-Mamba is a hybrid CNN-SSM architecture that outperforms prior CNN and Transformer networks on biomedical image segmentation tasks by efficiently modeling long-range dependencies.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the hybrid CNN-SSM block will reliably improve long-range dependency capture and generalization across diverse biomedical datasets without introducing training instability or requiring dataset-specific tuning beyond the claimed self-configuring mechanism.","pith_extraction_headline":"U-Mamba pairs convolutional layers with state space models to capture long-range dependencies more effectively than prior CNN or Transformer networks for biomedical image segmentation."},"references":{"count":49,"sample":[{"doi":"","year":2017,"title":"2017 robotic instrument segmentation challenge","work_id":"3ba7cf59-49b5-49a8-b685-9d61f96c94c6","ref_index":1,"cited_arxiv_id":"1902.06426","is_internal_anchor":true},{"doi":"","year":2016,"title":"Layer Normalization","work_id":"20a2d720-0046-4c7c-bcd6-327ec8143f69","ref_index":2,"cited_arxiv_id":"1607.06450","is_internal_anchor":true},{"doi":"","year":2023,"title":"Medical Image Analysis84, 102680 (2023) 2","work_id":"2d45f319-709b-41ef-9660-d293a5b04814","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"IEEE Transactions on Medical Imaging 40(12), 3543–3554 (2021) 9","work_id":"e6d85a46-0c89-43a0-aa25-55d63432e20c","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation","work_id":"ffa2ac60-2755-4390-9f66-07815aa6cb27","ref_index":5,"cited_arxiv_id":"2102.04306","is_internal_anchor":true}],"resolved_work":49,"snapshot_sha256":"86907e620988558993a6af55f8eec042fc2ba9eaa61e36adb6fc0e616943d3f4","internal_anchors":7},"formal_canon":{"evidence_count":3,"snapshot_sha256":"98d8300e6c1fd113ef69aaffc4dba39a1940aaaeb1fdaa7a64a4227877487275"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}