{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:YHZE32RRPA64JRCIGYFQDC2YQS","short_pith_number":"pith:YHZE32RR","schema_version":"1.0","canonical_sha256":"c1f24dea31783dc4c448360b018b5884898acc5fa73d76ec97b108f734a6ab7d","source":{"kind":"arxiv","id":"2103.12051","version":1},"attestation_state":"computed","paper":{"title":"SSD: A Unified Framework for Self-Supervised Outlier Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Mung Chiang, Prateek Mittal, Vikash Sehwag","submitted_at":"2021-03-22T17:51:35Z","abstract_excerpt":"We ask the following question: what training information is required to design an effective outlier/out-of-distribution (OOD) detector, i.e., detecting samples that lie far away from the training distribution? Since unlabeled data is easily accessible for many applications, the most compelling approach is to develop detectors based on only unlabeled in-distribution data. However, we observe that most existing detectors based on unlabeled data perform poorly, often equivalent to a random prediction. In contrast, existing state-of-the-art OOD detectors achieve impressive performance but require "},"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":"2103.12051","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-03-22T17:51:35Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"a55bf048dcc26635fd5f3931cfc6c3caa8f991107f37bd972f3fc58ab4858ac9","abstract_canon_sha256":"86bae1f053c96ad95b474a0cac8ae9dffbda8548574b291e5ebf2552a7a1b89b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:25:14.334767Z","signature_b64":"wNpDeZidi38D4soY9oKPlMYXPvbXG9rL81ncm1HHWq/x+qltsAc7dndk1aico7MCs1lMNCnfOKtMFT/thqRJDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c1f24dea31783dc4c448360b018b5884898acc5fa73d76ec97b108f734a6ab7d","last_reissued_at":"2026-07-05T02:25:14.334310Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:25:14.334310Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SSD: A Unified Framework for Self-Supervised Outlier Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Mung Chiang, Prateek Mittal, Vikash Sehwag","submitted_at":"2021-03-22T17:51:35Z","abstract_excerpt":"We ask the following question: what training information is required to design an effective outlier/out-of-distribution (OOD) detector, i.e., detecting samples that lie far away from the training distribution? Since unlabeled data is easily accessible for many applications, the most compelling approach is to develop detectors based on only unlabeled in-distribution data. However, we observe that most existing detectors based on unlabeled data perform poorly, often equivalent to a random prediction. In contrast, existing state-of-the-art OOD detectors achieve impressive performance but require "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.12051","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/2103.12051/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":"2103.12051","created_at":"2026-07-05T02:25:14.334370+00:00"},{"alias_kind":"arxiv_version","alias_value":"2103.12051v1","created_at":"2026-07-05T02:25:14.334370+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.12051","created_at":"2026-07-05T02:25:14.334370+00:00"},{"alias_kind":"pith_short_12","alias_value":"YHZE32RRPA64","created_at":"2026-07-05T02:25:14.334370+00:00"},{"alias_kind":"pith_short_16","alias_value":"YHZE32RRPA64JRCI","created_at":"2026-07-05T02:25:14.334370+00:00"},{"alias_kind":"pith_short_8","alias_value":"YHZE32RR","created_at":"2026-07-05T02:25:14.334370+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2501.08083","citing_title":"Benchmarking Vision Foundation Models for Input Monitoring in Autonomous Driving","ref_index":73,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17575","citing_title":"UniAlign: A Model-Agnostic Framework for Robust Network Traffic Classification under Distribution Shifts","ref_index":59,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10756","citing_title":"TINS: Test-time ID-prototype-separated Negative Semantics Learning for OOD Detection","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05776","citing_title":"HEDP: A Hybrid Energy-Distance Prompt-based Framework for Domain Incremental Learning","ref_index":59,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YHZE32RRPA64JRCIGYFQDC2YQS","json":"https://pith.science/pith/YHZE32RRPA64JRCIGYFQDC2YQS.json","graph_json":"https://pith.science/api/pith-number/YHZE32RRPA64JRCIGYFQDC2YQS/graph.json","events_json":"https://pith.science/api/pith-number/YHZE32RRPA64JRCIGYFQDC2YQS/events.json","paper":"https://pith.science/paper/YHZE32RR"},"agent_actions":{"view_html":"https://pith.science/pith/YHZE32RRPA64JRCIGYFQDC2YQS","download_json":"https://pith.science/pith/YHZE32RRPA64JRCIGYFQDC2YQS.json","view_paper":"https://pith.science/paper/YHZE32RR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2103.12051&json=true","fetch_graph":"https://pith.science/api/pith-number/YHZE32RRPA64JRCIGYFQDC2YQS/graph.json","fetch_events":"https://pith.science/api/pith-number/YHZE32RRPA64JRCIGYFQDC2YQS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YHZE32RRPA64JRCIGYFQDC2YQS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YHZE32RRPA64JRCIGYFQDC2YQS/action/storage_attestation","attest_author":"https://pith.science/pith/YHZE32RRPA64JRCIGYFQDC2YQS/action/author_attestation","sign_citation":"https://pith.science/pith/YHZE32RRPA64JRCIGYFQDC2YQS/action/citation_signature","submit_replication":"https://pith.science/pith/YHZE32RRPA64JRCIGYFQDC2YQS/action/replication_record"}},"created_at":"2026-07-05T02:25:14.334370+00:00","updated_at":"2026-07-05T02:25:14.334370+00:00"}