{"ID":5551784,"CreatedAt":"2026-07-02T01:54:51.863792489Z","UpdatedAt":"2026-07-04T09:21:41.829188432Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.00687","arxiv_id":"2607.00687","title":"LUMA: Benchmarking Segmentation via a Lightweight Universal Mask Adapter","abstract":"Comparing transformer backbones for image segmentation is confounded: each is paired with a different decoder, recipe, and pretraining, so reported differences rarely reflect the backbone itself. We introduce the Lightweight Universal Mask Adapter (LUMA), a lightweight, backbone-agnostic mask-transformer head that treats any backbone as a black-box feature extractor, letting a set of queries read from its features through cheap cross-attention. LUMA matches the accuracy of EoMT, the state-of-the-art efficient ViT-segmenter, at lower cost, while attaching unchanged to isotropic, hierarchical, convolutional, and mixture-of-experts backbones alike. Holding this head fixed, we benchmark 20 backbones, 11 pretraining schemes and a range of resolutions on ADE20K and Cityscapes under one modern recipe. We find that ``efficient'' token mixers fail to deliver efficiency even at the high resolutions that motivate them, with plain ViT holding the throughput Pareto-front at every resolution. Additionally, the pretraining objective, not the architecture, the lever the field has tuned hardest, governs segmentation quality.","short_abstract":"Comparing transformer backbones for image segmentation is confounded: each is paired with a different decoder, recipe, and pretraining, so reported differences rarely reflect the backbone itself. We introduce the Lightweight Universal Mask Adapter (LUMA), a lightweight, backbone-agnostic mask-transformer head that trea...","url_abs":"https://arxiv.org/abs/2607.00687","url_pdf":"https://arxiv.org/pdf/2607.00687v1","authors":"[\"Tobias Christian Nauen\",\"Anosh Billimoria\",\"Federico Raue\",\"Stanislav Frolov\",\"Brian B. Moser\",\"Andreas Dengel\"]","published":"2026-07-01T09:35:24Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.AI\",\"cs.LG\",\"cs.PF\"]","methods":"[\"Transformer\"]","has_code":false}