{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:KWJ2ZMNNSPEYC5KY5ZLYJCYJUR","short_pith_number":"pith:KWJ2ZMNN","schema_version":"1.0","canonical_sha256":"5593acb1ad93c9817558ee57848b09a451d854bd4bc2181302b05b8f0a4a2ffd","source":{"kind":"arxiv","id":"2402.00667","version":1},"attestation_state":"computed","paper":{"title":"Improving Weak-to-Strong Generalization with Scalable Oversight and Ensemble Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chao Kong, Jing Zhang, Jinlin Xiao, Jitao Sang, Junhong Ye, Shuyu Wei, Yanxu Zhu, Yuhang Wang","submitted_at":"2024-02-01T15:30:19Z","abstract_excerpt":"This paper presents a follow-up study to OpenAI's recent superalignment work on Weak-to-Strong Generalization (W2SG). Superalignment focuses on ensuring that high-level AI systems remain consistent with human values and intentions when dealing with complex, high-risk tasks. The W2SG framework has opened new possibilities for empirical research in this evolving field. Our study simulates two phases of superalignment under the W2SG framework: the development of general superhuman models and the progression towards superintelligence. In the first phase, based on human supervision, the quality of "},"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":"2402.00667","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-01T15:30:19Z","cross_cats_sorted":[],"title_canon_sha256":"3a6daa83ead4939bbbdb8cd9c55648db18deaf8abc72bcb8d3a04ceeb4d78b3b","abstract_canon_sha256":"331ce50166611a78c5e8d3c4a7d0e2689b48156f33a5bfc71e53da7bef90a3fa"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:40:16.254258Z","signature_b64":"rtCRh+SZoUD85Olap7D/Hztp54zOPqa8sK2475W344+0UXMkMm2/a5EEIdLsR/hJ8xznxLsT3UnrH2Kzud66Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5593acb1ad93c9817558ee57848b09a451d854bd4bc2181302b05b8f0a4a2ffd","last_reissued_at":"2026-07-05T07:40:16.253773Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:40:16.253773Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving Weak-to-Strong Generalization with Scalable Oversight and Ensemble Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chao Kong, Jing Zhang, Jinlin Xiao, Jitao Sang, Junhong Ye, Shuyu Wei, Yanxu Zhu, Yuhang Wang","submitted_at":"2024-02-01T15:30:19Z","abstract_excerpt":"This paper presents a follow-up study to OpenAI's recent superalignment work on Weak-to-Strong Generalization (W2SG). Superalignment focuses on ensuring that high-level AI systems remain consistent with human values and intentions when dealing with complex, high-risk tasks. The W2SG framework has opened new possibilities for empirical research in this evolving field. Our study simulates two phases of superalignment under the W2SG framework: the development of general superhuman models and the progression towards superintelligence. In the first phase, based on human supervision, the quality of "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.00667","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/2402.00667/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":"2402.00667","created_at":"2026-07-05T07:40:16.253831+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.00667v1","created_at":"2026-07-05T07:40:16.253831+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.00667","created_at":"2026-07-05T07:40:16.253831+00:00"},{"alias_kind":"pith_short_12","alias_value":"KWJ2ZMNNSPEY","created_at":"2026-07-05T07:40:16.253831+00:00"},{"alias_kind":"pith_short_16","alias_value":"KWJ2ZMNNSPEYC5KY","created_at":"2026-07-05T07:40:16.253831+00:00"},{"alias_kind":"pith_short_8","alias_value":"KWJ2ZMNN","created_at":"2026-07-05T07:40:16.253831+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05394","citing_title":"Weak-to-Strong Generalization via Direct On-Policy Distillation","ref_index":95,"is_internal_anchor":true},{"citing_arxiv_id":"2505.03631","citing_title":"Generalizable Video Quality Assessment via Weak-to-Strong Learning","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05710","citing_title":"On the Blessing of Pre-training in Weak-to-Strong Generalization","ref_index":118,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KWJ2ZMNNSPEYC5KY5ZLYJCYJUR","json":"https://pith.science/pith/KWJ2ZMNNSPEYC5KY5ZLYJCYJUR.json","graph_json":"https://pith.science/api/pith-number/KWJ2ZMNNSPEYC5KY5ZLYJCYJUR/graph.json","events_json":"https://pith.science/api/pith-number/KWJ2ZMNNSPEYC5KY5ZLYJCYJUR/events.json","paper":"https://pith.science/paper/KWJ2ZMNN"},"agent_actions":{"view_html":"https://pith.science/pith/KWJ2ZMNNSPEYC5KY5ZLYJCYJUR","download_json":"https://pith.science/pith/KWJ2ZMNNSPEYC5KY5ZLYJCYJUR.json","view_paper":"https://pith.science/paper/KWJ2ZMNN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.00667&json=true","fetch_graph":"https://pith.science/api/pith-number/KWJ2ZMNNSPEYC5KY5ZLYJCYJUR/graph.json","fetch_events":"https://pith.science/api/pith-number/KWJ2ZMNNSPEYC5KY5ZLYJCYJUR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KWJ2ZMNNSPEYC5KY5ZLYJCYJUR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KWJ2ZMNNSPEYC5KY5ZLYJCYJUR/action/storage_attestation","attest_author":"https://pith.science/pith/KWJ2ZMNNSPEYC5KY5ZLYJCYJUR/action/author_attestation","sign_citation":"https://pith.science/pith/KWJ2ZMNNSPEYC5KY5ZLYJCYJUR/action/citation_signature","submit_replication":"https://pith.science/pith/KWJ2ZMNNSPEYC5KY5ZLYJCYJUR/action/replication_record"}},"created_at":"2026-07-05T07:40:16.253831+00:00","updated_at":"2026-07-05T07:40:16.253831+00:00"}