{"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/v-kd-improving-knowledge-distillation-using","title":"$V_kD:$ Improving Knowledge Distillation using Orthogonal Projections","arxiv_id":"2403.06213","date":"2024-03-10","proceeding":null,"authors":["Roy Miles","Ismail Elezi","Jiankang Deng"],"abstract":"Knowledge distillation is an effective method for training small and efficient deep learning models. However, the efficacy of a single method can degenerate when transferring to other tasks, modalities, or even other architectures. To address this limitation, we propose a novel constrained feature distillation method. This method is derived from a small set of core principles, which results in two emerging components: an orthogonal projection and a task-specific normalisation. Equipped with both of these components, our transformer models can outperform all previous methods on ImageNet and reach up to a 4.4% relative improvement over the previous state-of-the-art methods. To further demonstrate the generality of our method, we apply it to object detection and image generation, whereby we obtain consistent and substantial performance improvements over state-of-the-art. Code and models are publicly available: https://github.com/roymiles/vkd","url_abs":"https://arxiv.org/abs/2403.06213v1","url_pdf":"https://arxiv.org/pdf/2403.06213v1.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":"v-kd-improving-knowledge-distillation-using","repo_url":"https://github.com/roymiles/vkd","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"image-generation","task_name":"Image Generation"},{"task_slug":"knowledge-distillation","task_name":"Knowledge Distillation"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[{"method_slug":"set","method_name":"SET"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/knowledge-distillation-on-imagenet","task":"Knowledge Distillation","dataset":"ImageNet","model":"VkD (T:RegNety 160 S:DeiT-S)","rank_in_archive_order":6,"of":52,"metrics":{"CRD training setting":"✘","Top-1 accuracy %":"82.9","model size":"22M"},"uses_additional_data":false},{"leaderboard":"/sota/knowledge-distillation-on-imagenet","task":"Knowledge Distillation","dataset":"ImageNet","model":"VkD (T:RegNety 160 S:DeiT-Ti)","rank_in_archive_order":15,"of":52,"metrics":{"CRD training setting":"✘","Top-1 accuracy %":"79.2","model size":"6M"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2403.06213","atlas_url":"https://app.syntology.ai/?focus=2403.06213","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2403.06213"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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/roymiles/vkd","reach":null}],"summary":{"ran_draft_wrong":2,"ran_fixture":2,"unverified":1},"by_repo_kind":{"official":{"samples":5,"ran":4,"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":5,"samples":[{"code_sha256_prefix":"896fc4d5053e27de","entry":"dice_loss","repo":"roymiles/vkd","repo_kind":"official","path":"vidt/methods/vidt/criterion.py","file_url":"https://github.com/roymiles/vkd/blob/HEAD/vidt/methods/vidt/criterion.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"896fc4d5053e27de"}},{"code_sha256_prefix":"dc8fb9b6fb48ac37","entry":"generalized_box_iou","repo":"roymiles/vkd","repo_kind":"official","path":"vidt/methods/vidt/criterion.py","file_url":"https://github.com/roymiles/vkd/blob/HEAD/vidt/methods/vidt/criterion.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"dc8fb9b6fb48ac37"}},{"code_sha256_prefix":"a818498638fbbcce","entry":"interpolate","repo":"roymiles/vkd","repo_kind":"official","path":"vidt/methods/vidt/criterion.py","file_url":"https://github.com/roymiles/vkd/blob/HEAD/vidt/methods/vidt/criterion.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"a818498638fbbcce"}},{"code_sha256_prefix":"bf3a75e08dc2c405","entry":"sigmoid_focal_loss","repo":"roymiles/vkd","repo_kind":"official","path":"vidt/methods/vidt/criterion.py","file_url":"https://github.com/roymiles/vkd/blob/HEAD/vidt/methods/vidt/criterion.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"bf3a75e08dc2c405"}},{"code_sha256_prefix":"cad80663e56e3c1e","entry":"SetCriterion","repo":"roymiles/vkd","repo_kind":"official","path":"vidt/methods/vidt/criterion.py","file_url":"https://github.com/roymiles/vkd/blob/HEAD/vidt/methods/vidt/criterion.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"cad80663e56e3c1e"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}