{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:5LZMK2ZJ53X2JNSTZNCUFVX3RU","short_pith_number":"pith:5LZMK2ZJ","schema_version":"1.0","canonical_sha256":"eaf2c56b29eeefa4b653cb4542d6fb8d17e2fbfd5a5bde778dc3fd9dc003843e","source":{"kind":"arxiv","id":"2108.06682","version":1},"attestation_state":"computed","paper":{"title":"ST3D++: Denoised Self-training for Unsupervised Domain Adaptation on 3D Object Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Hongsheng Li, Jihan Yang, Shaoshuai Shi, Xiaojuan Qi, Zhe Wang","submitted_at":"2021-08-15T07:49:06Z","abstract_excerpt":"In this paper, we present a self-training method, named ST3D++, with a holistic pseudo label denoising pipeline for unsupervised domain adaptation on 3D object detection. ST3D++ aims at reducing noise in pseudo label generation as well as alleviating the negative impacts of noisy pseudo labels on model training. First, ST3D++ pre-trains the 3D object detector on the labeled source domain with random object scaling (ROS) which is designed to reduce target domain pseudo label noise arising from object scale bias of the source domain. Then, the detector is progressively improved through alternati"},"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":"2108.06682","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-08-15T07:49:06Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"eb51408f3f45f2cb51d7db4a34c388daeaf10fdd2f1230abac0168fea052f769","abstract_canon_sha256":"61fdb4cd296d3f0956bcd77832a74d4ed256308b022e1d37d254c2ca45e117e4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:07:16.356841Z","signature_b64":"+8Z3FFCWw9koMm29oLZM8Ga3VFRiUtlh1z1gIyz7bGZbK7V+zdEZdi6Q8RVQ93+5gzDluqHvZlRYxaGJR1TVBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"eaf2c56b29eeefa4b653cb4542d6fb8d17e2fbfd5a5bde778dc3fd9dc003843e","last_reissued_at":"2026-07-05T03:07:16.356288Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:07:16.356288Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ST3D++: Denoised Self-training for Unsupervised Domain Adaptation on 3D Object Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Hongsheng Li, Jihan Yang, Shaoshuai Shi, Xiaojuan Qi, Zhe Wang","submitted_at":"2021-08-15T07:49:06Z","abstract_excerpt":"In this paper, we present a self-training method, named ST3D++, with a holistic pseudo label denoising pipeline for unsupervised domain adaptation on 3D object detection. ST3D++ aims at reducing noise in pseudo label generation as well as alleviating the negative impacts of noisy pseudo labels on model training. First, ST3D++ pre-trains the 3D object detector on the labeled source domain with random object scaling (ROS) which is designed to reduce target domain pseudo label noise arising from object scale bias of the source domain. Then, the detector is progressively improved through alternati"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2108.06682","kind":"arxiv","version":1},"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/2108.06682/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":"2108.06682","created_at":"2026-07-05T03:07:16.356383+00:00"},{"alias_kind":"arxiv_version","alias_value":"2108.06682v1","created_at":"2026-07-05T03:07:16.356383+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2108.06682","created_at":"2026-07-05T03:07:16.356383+00:00"},{"alias_kind":"pith_short_12","alias_value":"5LZMK2ZJ53X2","created_at":"2026-07-05T03:07:16.356383+00:00"},{"alias_kind":"pith_short_16","alias_value":"5LZMK2ZJ53X2JNST","created_at":"2026-07-05T03:07:16.356383+00:00"},{"alias_kind":"pith_short_8","alias_value":"5LZMK2ZJ","created_at":"2026-07-05T03:07:16.356383+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.04950","citing_title":"MORDA: A Synthetic Dataset to Facilitate Adaptation of Object Detectors to Unseen Real-target Domain While Preserving Performance on Real-source Domain","ref_index":12,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5LZMK2ZJ53X2JNSTZNCUFVX3RU","json":"https://pith.science/pith/5LZMK2ZJ53X2JNSTZNCUFVX3RU.json","graph_json":"https://pith.science/api/pith-number/5LZMK2ZJ53X2JNSTZNCUFVX3RU/graph.json","events_json":"https://pith.science/api/pith-number/5LZMK2ZJ53X2JNSTZNCUFVX3RU/events.json","paper":"https://pith.science/paper/5LZMK2ZJ"},"agent_actions":{"view_html":"https://pith.science/pith/5LZMK2ZJ53X2JNSTZNCUFVX3RU","download_json":"https://pith.science/pith/5LZMK2ZJ53X2JNSTZNCUFVX3RU.json","view_paper":"https://pith.science/paper/5LZMK2ZJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2108.06682&json=true","fetch_graph":"https://pith.science/api/pith-number/5LZMK2ZJ53X2JNSTZNCUFVX3RU/graph.json","fetch_events":"https://pith.science/api/pith-number/5LZMK2ZJ53X2JNSTZNCUFVX3RU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5LZMK2ZJ53X2JNSTZNCUFVX3RU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5LZMK2ZJ53X2JNSTZNCUFVX3RU/action/storage_attestation","attest_author":"https://pith.science/pith/5LZMK2ZJ53X2JNSTZNCUFVX3RU/action/author_attestation","sign_citation":"https://pith.science/pith/5LZMK2ZJ53X2JNSTZNCUFVX3RU/action/citation_signature","submit_replication":"https://pith.science/pith/5LZMK2ZJ53X2JNSTZNCUFVX3RU/action/replication_record"}},"created_at":"2026-07-05T03:07:16.356383+00:00","updated_at":"2026-07-05T03:07:16.356383+00:00"}