{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:IBWVE6N4CO5J3PYHO4BTX4ZBXW","short_pith_number":"pith:IBWVE6N4","schema_version":"1.0","canonical_sha256":"406d5279bc13ba9dbf0777033bf321bd82ef99fed49a4e0b0b9d9424e8db884b","source":{"kind":"arxiv","id":"2203.01305","version":3},"attestation_state":"computed","paper":{"title":"DN-DETR: Accelerate DETR Training by Introducing Query DeNoising","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Feng Li, Hao Zhang, Jian Guo, Lei Zhang, Lionel M. Ni, Shilong Liu","submitted_at":"2022-03-02T18:50:23Z","abstract_excerpt":"We present in this paper a novel denoising training method to speedup DETR (DEtection TRansformer) training and offer a deepened understanding of the slow convergence issue of DETR-like methods. We show that the slow convergence results from the instability of bipartite graph matching which causes inconsistent optimization goals in early training stages. To address this issue, except for the Hungarian loss, our method additionally feeds ground-truth bounding boxes with noises into Transformer decoder and trains the model to reconstruct the original boxes, which effectively reduces the bipartit"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2203.01305","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-03-02T18:50:23Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"32289e53131875c3ea018b532f39920ef3026558a3a93fe1c39009892f783b0d","abstract_canon_sha256":"91190fc0519e143167418459836fb74ab900df18701af4fc869d17aa6459e051"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:23:29.863016Z","signature_b64":"uUwXmhXn4BugdeEOpyvLOa6DZGu5Cv1eNcuYI9hTlSQto/gAdvXzjFkPj1JihnkkRvi71Cgm2VhIVOUiD4iWBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"406d5279bc13ba9dbf0777033bf321bd82ef99fed49a4e0b0b9d9424e8db884b","last_reissued_at":"2026-07-05T05:23:29.862513Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:23:29.862513Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DN-DETR: Accelerate DETR Training by Introducing Query DeNoising","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Feng Li, Hao Zhang, Jian Guo, Lei Zhang, Lionel M. Ni, Shilong Liu","submitted_at":"2022-03-02T18:50:23Z","abstract_excerpt":"We present in this paper a novel denoising training method to speedup DETR (DEtection TRansformer) training and offer a deepened understanding of the slow convergence issue of DETR-like methods. We show that the slow convergence results from the instability of bipartite graph matching which causes inconsistent optimization goals in early training stages. To address this issue, except for the Hungarian loss, our method additionally feeds ground-truth bounding boxes with noises into Transformer decoder and trains the model to reconstruct the original boxes, which effectively reduces the bipartit"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.01305","kind":"arxiv","version":3},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2203.01305/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2203.01305","created_at":"2026-07-05T05:23:29.862575+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.01305v3","created_at":"2026-07-05T05:23:29.862575+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.01305","created_at":"2026-07-05T05:23:29.862575+00:00"},{"alias_kind":"pith_short_12","alias_value":"IBWVE6N4CO5J","created_at":"2026-07-05T05:23:29.862575+00:00"},{"alias_kind":"pith_short_16","alias_value":"IBWVE6N4CO5J3PYH","created_at":"2026-07-05T05:23:29.862575+00:00"},{"alias_kind":"pith_short_8","alias_value":"IBWVE6N4","created_at":"2026-07-05T05:23:29.862575+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.08004","citing_title":"LOGOS: Language-guided Oriented Object Detection in Aerial Scenes","ref_index":15,"is_internal_anchor":true},{"citing_arxiv_id":"2607.00385","citing_title":"MalariAI: A Label-Resilient Decoupled Framework for Annotation-Agnostic Cell Segmentation and Explainable Stage Classification in Dense Malaria Blood Smears","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2203.03605","citing_title":"DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detection","ref_index":17,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IBWVE6N4CO5J3PYHO4BTX4ZBXW","json":"https://pith.science/pith/IBWVE6N4CO5J3PYHO4BTX4ZBXW.json","graph_json":"https://pith.science/api/pith-number/IBWVE6N4CO5J3PYHO4BTX4ZBXW/graph.json","events_json":"https://pith.science/api/pith-number/IBWVE6N4CO5J3PYHO4BTX4ZBXW/events.json","paper":"https://pith.science/paper/IBWVE6N4"},"agent_actions":{"view_html":"https://pith.science/pith/IBWVE6N4CO5J3PYHO4BTX4ZBXW","download_json":"https://pith.science/pith/IBWVE6N4CO5J3PYHO4BTX4ZBXW.json","view_paper":"https://pith.science/paper/IBWVE6N4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.01305&json=true","fetch_graph":"https://pith.science/api/pith-number/IBWVE6N4CO5J3PYHO4BTX4ZBXW/graph.json","fetch_events":"https://pith.science/api/pith-number/IBWVE6N4CO5J3PYHO4BTX4ZBXW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IBWVE6N4CO5J3PYHO4BTX4ZBXW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IBWVE6N4CO5J3PYHO4BTX4ZBXW/action/storage_attestation","attest_author":"https://pith.science/pith/IBWVE6N4CO5J3PYHO4BTX4ZBXW/action/author_attestation","sign_citation":"https://pith.science/pith/IBWVE6N4CO5J3PYHO4BTX4ZBXW/action/citation_signature","submit_replication":"https://pith.science/pith/IBWVE6N4CO5J3PYHO4BTX4ZBXW/action/replication_record"}},"created_at":"2026-07-05T05:23:29.862575+00:00","updated_at":"2026-07-05T05:23:29.862575+00:00"}