{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:2O555ZYSRTCISMB6CMEUDKVUBI","short_pith_number":"pith:2O555ZYS","schema_version":"1.0","canonical_sha256":"d3bbdee7128cc489303e130941aab40a22e97fbf9b34915999decfddb07fcc59","source":{"kind":"arxiv","id":"2210.10019","version":4},"attestation_state":"computed","paper":{"title":"Towards Understanding GD with Hard and Conjugate Pseudo-labels for Test-Time Adaptation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Andre Wibisono, Jun-Kun Wang","submitted_at":"2022-10-18T17:39:32Z","abstract_excerpt":"We consider a setting that a model needs to adapt to a new domain under distribution shifts, given that only unlabeled test samples from the new domain are accessible at test time. A common idea in most of the related works is constructing pseudo-labels for the unlabeled test samples and applying gradient descent (GD) to a loss function with the pseudo-labels. Recently, \\cite{GSRK22} propose conjugate labels, which is a new kind of pseudo-labels for self-training at test time. They empirically show that the conjugate label outperforms other ways of pseudo-labeling on many domain adaptation ben"},"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":"2210.10019","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-10-18T17:39:32Z","cross_cats_sorted":[],"title_canon_sha256":"9f3c1f3d65dacfe2b9a2799b0ed26c8d9aa284ab6f31f3d38b031c5c9522c06f","abstract_canon_sha256":"9667871e4666b08df9fe38b29cae82deaf2c7202ded2f7ec62b7c1a705907c37"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:45:33.907638Z","signature_b64":"IMwJyVrSv5PqZh+DnWk8KTH6eRVamXqksTXph/bZTsuspypIDjBinbXRMmtmyfZf+kuQqKrIVYFYEogbikytAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d3bbdee7128cc489303e130941aab40a22e97fbf9b34915999decfddb07fcc59","last_reissued_at":"2026-07-05T05:45:33.907176Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:45:33.907176Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Understanding GD with Hard and Conjugate Pseudo-labels for Test-Time Adaptation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Andre Wibisono, Jun-Kun Wang","submitted_at":"2022-10-18T17:39:32Z","abstract_excerpt":"We consider a setting that a model needs to adapt to a new domain under distribution shifts, given that only unlabeled test samples from the new domain are accessible at test time. A common idea in most of the related works is constructing pseudo-labels for the unlabeled test samples and applying gradient descent (GD) to a loss function with the pseudo-labels. Recently, \\cite{GSRK22} propose conjugate labels, which is a new kind of pseudo-labels for self-training at test time. They empirically show that the conjugate label outperforms other ways of pseudo-labeling on many domain adaptation ben"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.10019","kind":"arxiv","version":4},"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/2210.10019/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":"2210.10019","created_at":"2026-07-05T05:45:33.907235+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.10019v4","created_at":"2026-07-05T05:45:33.907235+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.10019","created_at":"2026-07-05T05:45:33.907235+00:00"},{"alias_kind":"pith_short_12","alias_value":"2O555ZYSRTCI","created_at":"2026-07-05T05:45:33.907235+00:00"},{"alias_kind":"pith_short_16","alias_value":"2O555ZYSRTCISMB6","created_at":"2026-07-05T05:45:33.907235+00:00"},{"alias_kind":"pith_short_8","alias_value":"2O555ZYS","created_at":"2026-07-05T05:45:33.907235+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.18337","citing_title":"DART$^3$: Leveraging Distance for Test Time Adaptation in Person Re-Identification","ref_index":26,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2O555ZYSRTCISMB6CMEUDKVUBI","json":"https://pith.science/pith/2O555ZYSRTCISMB6CMEUDKVUBI.json","graph_json":"https://pith.science/api/pith-number/2O555ZYSRTCISMB6CMEUDKVUBI/graph.json","events_json":"https://pith.science/api/pith-number/2O555ZYSRTCISMB6CMEUDKVUBI/events.json","paper":"https://pith.science/paper/2O555ZYS"},"agent_actions":{"view_html":"https://pith.science/pith/2O555ZYSRTCISMB6CMEUDKVUBI","download_json":"https://pith.science/pith/2O555ZYSRTCISMB6CMEUDKVUBI.json","view_paper":"https://pith.science/paper/2O555ZYS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.10019&json=true","fetch_graph":"https://pith.science/api/pith-number/2O555ZYSRTCISMB6CMEUDKVUBI/graph.json","fetch_events":"https://pith.science/api/pith-number/2O555ZYSRTCISMB6CMEUDKVUBI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2O555ZYSRTCISMB6CMEUDKVUBI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2O555ZYSRTCISMB6CMEUDKVUBI/action/storage_attestation","attest_author":"https://pith.science/pith/2O555ZYSRTCISMB6CMEUDKVUBI/action/author_attestation","sign_citation":"https://pith.science/pith/2O555ZYSRTCISMB6CMEUDKVUBI/action/citation_signature","submit_replication":"https://pith.science/pith/2O555ZYSRTCISMB6CMEUDKVUBI/action/replication_record"}},"created_at":"2026-07-05T05:45:33.907235+00:00","updated_at":"2026-07-05T05:45:33.907235+00:00"}