{"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/aligning-generative-denoising-with","title":"Aligning Generative Denoising with Discriminative Objectives Unleashes Diffusion for Visual Perception","arxiv_id":"2504.11457","date":"2025-04-15","proceeding":null,"authors":["Ziqi Pang","Xin Xu","Yu-Xiong Wang"],"abstract":"With the success of image generation, generative diffusion models are increasingly adopted for discriminative tasks, as pixel generation provides a unified perception interface. However, directly repurposing the generative denoising process for discriminative objectives reveals critical gaps rarely addressed previously. Generative models tolerate intermediate sampling errors if the final distribution remains plausible, but discriminative tasks require rigorous accuracy throughout, as evidenced in challenging multi-modal tasks like referring image segmentation. Motivated by this gap, we analyze and enhance alignment between generative diffusion processes and perception tasks, focusing on how perception quality evolves during denoising. We find: (1) earlier denoising steps contribute disproportionately to perception quality, prompting us to propose tailored learning objectives reflecting varying timestep contributions; (2) later denoising steps show unexpected perception degradation, highlighting sensitivity to training-denoising distribution shifts, addressed by our diffusion-tailored data augmentation; and (3) generative processes uniquely enable interactivity, serving as controllable user interfaces adaptable to correctional prompts in multi-round interactions. Our insights significantly improve diffusion-based perception models without architectural changes, achieving state-of-the-art performance on depth estimation, referring image segmentation, and generalist perception tasks. Code available at https://github.com/ziqipang/ADDP.","url_abs":"https://arxiv.org/abs/2504.11457v1","url_pdf":"https://arxiv.org/pdf/2504.11457v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"aligning-generative-denoising-with","repo_url":"https://github.com/ziqipang/addp","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"data-augmentation","task_name":"Data Augmentation"},{"task_slug":"denoising","task_name":"Denoising"},{"task_slug":"depth-estimation","task_name":"Depth Estimation"},{"task_slug":"image-generation","task_name":"Image Generation"},{"task_slug":"image-segmentation","task_name":"Image Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"}],"methods":[{"method_slug":"diffusion","method_name":"Diffusion"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2504.11457","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2504.11457"}},"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":[{"provenance":"deterministic:regex_extraction","url":"https://github.com/ziqipang/ADDP","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ziqipang/addp","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_violates":1,"ran_draft_wrong":1,"unverified":1},"by_repo_kind":{"official":{"samples":3,"ran":2,"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":0,"samples":[{"code_sha256_prefix":"4cb732f513d69dfd","entry":"disabled_train","repo":"ziqipang/ADDP","repo_kind":"official","path":"referring_segmentation/stable_diffusion/ldm/models/diffusion/ddpm_edit_addp.py","file_url":"https://github.com/ziqipang/ADDP/blob/HEAD/referring_segmentation/stable_diffusion/ldm/models/diffusion/ddpm_edit_addp.py","link_basis":"harvester_set","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4cb732f513d69dfd"}},{"code_sha256_prefix":"d48d8354986e3b0e","entry":"uniform_on_device","repo":"ziqipang/ADDP","repo_kind":"official","path":"referring_segmentation/stable_diffusion/ldm/models/diffusion/ddpm_edit_addp.py","file_url":"https://github.com/ziqipang/ADDP/blob/HEAD/referring_segmentation/stable_diffusion/ldm/models/diffusion/ddpm_edit_addp.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"d48d8354986e3b0e"}},{"code_sha256_prefix":"19b4e7dbd2c70b35","entry":"get_pred_name","repo":"ziqipang/ADDP","repo_kind":"official","path":"depth_estimation/src/dataset/base_depth_dataset.py","file_url":"https://github.com/ziqipang/ADDP/blob/HEAD/depth_estimation/src/dataset/base_depth_dataset.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"19b4e7dbd2c70b35"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}