{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/arxiv-2510-22229","title":"Diffusion-Driven Two-Stage Active Learning for Low-Budget Semantic Segmentation","arxiv_id":"2510.22229","date":"2025-10-25","proceeding":"NeurIPS","authors":["Jeongin Kim","Wonho Bae","YouLee Han","Giyeong Oh","Youngjae Yu","Danica J. Sutherland","Junhyug Noh"],"abstract":"Semantic segmentation demands dense pixel-level annotations, which can be prohibitively expensive - especially under extremely constrained labeling budgets. In this paper, we address the problem of low-budget active learning for semantic segmentation by proposing a novel two-stage selection pipeline. Our approach leverages a pre-trained diffusion model to extract rich multi-scale features that capture both global structure and fine details. In the first stage, we perform a hierarchical, representation-based candidate selection by first choosing a small subset of representative pixels per image using MaxHerding, and then refining these into a diverse global pool. In the second stage, we compute an entropy-augmented disagreement score (eDALD) over noisy multi-scale diffusion features to capture both epistemic uncertainty and prediction confidence, selecting the most informative pixels for annotation. This decoupling of diversity and uncertainty lets us achieve high segmentation accuracy with only a tiny fraction of labeled pixels. Extensive experiments on four benchmarks (CamVid, ADE-Bed, Cityscapes, and Pascal-Context) demonstrate that our method significantly outperforms existing baselines under extreme pixel-budget regimes. Our code is available at https://github.com/jn-kim/two-stage-edald.","url_abs":"https://arxiv.org/abs/2510.22229","url_pdf":"https://arxiv.org/pdf/2510.22229","source":{"archive":null,"snapshot":"2025-07-28","note":"not in the Papers with Code archive (frozen at the snapshot)","row_kind":"graph","title_abstract_authors_date":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)"},"code_links":[],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2510.22229","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2510.22229"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"mentioned_in_github":null,"is_official":null,"provenance":"deterministic:regex_extraction","mentioned_in_paper":null,"url":"https://github.com/jn-kim/two-stage-edald","reach":null}],"summary":{"ran":1,"unverified":3},"by_repo_kind":{"found_in_text":{"samples":4,"ran":1,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":4,"samples":[{"code_sha256_prefix":"00428bab72860435","entry":"UncertaintySampler","repo":"jn-kim/two-stage-edald","repo_kind":"found_in_text","path":"utils/query_utils.py","file_url":"https://github.com/jn-kim/two-stage-edald/blob/HEAD/utils/query_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"00428bab72860435"}},{"code_sha256_prefix":"fbcd3d55825e0572","entry":"bald_acquisition_function","repo":"jn-kim/two-stage-edald","repo_kind":"found_in_text","path":"utils/query_utils.py","file_url":"https://github.com/jn-kim/two-stage-edald/blob/HEAD/utils/query_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"fbcd3d55825e0572"}},{"code_sha256_prefix":"b5db073c06e97592","entry":"balentacq_acquisition_function","repo":"jn-kim/two-stage-edald","repo_kind":"found_in_text","path":"utils/query_utils.py","file_url":"https://github.com/jn-kim/two-stage-edald/blob/HEAD/utils/query_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"b5db073c06e97592"}},{"code_sha256_prefix":"c339159f0714913b","entry":"power_bald_acquisition_function","repo":"jn-kim/two-stage-edald","repo_kind":"found_in_text","path":"utils/query_utils.py","file_url":"https://github.com/jn-kim/two-stage-edald/blob/HEAD/utils/query_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"c339159f0714913b"}}]},"arxiv_metadata":{"licence":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)","fields":["title","abstract","authors","date"],"primary_category":"cs.CV","source":"arxiv_api"},"syntology_extracted_results":null}