{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:IVEWIKETXVP2PMKMK73DKMUMGJ","short_pith_number":"pith:IVEWIKET","schema_version":"1.0","canonical_sha256":"4549642893bd5fa7b14c57f635328c32613412970da88c699b1c570b91701bc8","source":{"kind":"arxiv","id":"2402.01909","version":2},"attestation_state":"computed","paper":{"title":"On Catastrophic Inheritance of Large Foundation Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CY"],"primary_cat":"cs.LG","authors_text":"Bhiksha Raj, Hao Chen, Jindong Wang, Xing Xie","submitted_at":"2024-02-02T21:21:55Z","abstract_excerpt":"Large foundation models (LFMs) are claiming incredible performances. Yet great concerns have been raised about their mythic and uninterpreted potentials not only in machine learning, but also in various other disciplines. In this position paper, we propose to identify a neglected issue deeply rooted in LFMs: Catastrophic Inheritance, describing the weaknesses and limitations inherited from biased large-scale pre-training data to behaviors of LFMs on the downstream tasks, including samples that are corrupted, long-tailed, noisy, out-of-distributed, to name a few. Such inheritance can potentiall"},"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":"2402.01909","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-02T21:21:55Z","cross_cats_sorted":["cs.AI","cs.CY"],"title_canon_sha256":"3e81affd47bf0708ed615a6de0da6ba8921409ee7c1a8dc58dae11af55a1eebf","abstract_canon_sha256":"70f7ef91d5667534881f7e3996a27291c61db1f082aa3f5bc719b5374932902b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:24:15.887361Z","signature_b64":"dajOGe2Nx0f6t8fYvt6x+qh3N/aItzffYqNiEx03RsjmRGNPLetcoENhlCIqoluGjQ1Gtm/MlvervbZY3aBVBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4549642893bd5fa7b14c57f635328c32613412970da88c699b1c570b91701bc8","last_reissued_at":"2026-07-05T09:24:15.886898Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:24:15.886898Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On Catastrophic Inheritance of Large Foundation Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CY"],"primary_cat":"cs.LG","authors_text":"Bhiksha Raj, Hao Chen, Jindong Wang, Xing Xie","submitted_at":"2024-02-02T21:21:55Z","abstract_excerpt":"Large foundation models (LFMs) are claiming incredible performances. Yet great concerns have been raised about their mythic and uninterpreted potentials not only in machine learning, but also in various other disciplines. In this position paper, we propose to identify a neglected issue deeply rooted in LFMs: Catastrophic Inheritance, describing the weaknesses and limitations inherited from biased large-scale pre-training data to behaviors of LFMs on the downstream tasks, including samples that are corrupted, long-tailed, noisy, out-of-distributed, to name a few. Such inheritance can potentiall"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.01909","kind":"arxiv","version":2},"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/2402.01909/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":"2402.01909","created_at":"2026-07-05T09:24:15.886952+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.01909v2","created_at":"2026-07-05T09:24:15.886952+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.01909","created_at":"2026-07-05T09:24:15.886952+00:00"},{"alias_kind":"pith_short_12","alias_value":"IVEWIKETXVP2","created_at":"2026-07-05T09:24:15.886952+00:00"},{"alias_kind":"pith_short_16","alias_value":"IVEWIKETXVP2PMKM","created_at":"2026-07-05T09:24:15.886952+00:00"},{"alias_kind":"pith_short_8","alias_value":"IVEWIKET","created_at":"2026-07-05T09:24:15.886952+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2405.12969","citing_title":"EchoAlign: Bridging Generative and Discriminative Learning under Noisy Labels","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20366","citing_title":"Mitigating Hallucinations in Large Vision-Language Models without Performance Degradation","ref_index":117,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IVEWIKETXVP2PMKMK73DKMUMGJ","json":"https://pith.science/pith/IVEWIKETXVP2PMKMK73DKMUMGJ.json","graph_json":"https://pith.science/api/pith-number/IVEWIKETXVP2PMKMK73DKMUMGJ/graph.json","events_json":"https://pith.science/api/pith-number/IVEWIKETXVP2PMKMK73DKMUMGJ/events.json","paper":"https://pith.science/paper/IVEWIKET"},"agent_actions":{"view_html":"https://pith.science/pith/IVEWIKETXVP2PMKMK73DKMUMGJ","download_json":"https://pith.science/pith/IVEWIKETXVP2PMKMK73DKMUMGJ.json","view_paper":"https://pith.science/paper/IVEWIKET","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.01909&json=true","fetch_graph":"https://pith.science/api/pith-number/IVEWIKETXVP2PMKMK73DKMUMGJ/graph.json","fetch_events":"https://pith.science/api/pith-number/IVEWIKETXVP2PMKMK73DKMUMGJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IVEWIKETXVP2PMKMK73DKMUMGJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IVEWIKETXVP2PMKMK73DKMUMGJ/action/storage_attestation","attest_author":"https://pith.science/pith/IVEWIKETXVP2PMKMK73DKMUMGJ/action/author_attestation","sign_citation":"https://pith.science/pith/IVEWIKETXVP2PMKMK73DKMUMGJ/action/citation_signature","submit_replication":"https://pith.science/pith/IVEWIKETXVP2PMKMK73DKMUMGJ/action/replication_record"}},"created_at":"2026-07-05T09:24:15.886952+00:00","updated_at":"2026-07-05T09:24:15.886952+00:00"}