{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:KXTRWL3JLN7V64OWYZ7GSOB3FJ","short_pith_number":"pith:KXTRWL3J","schema_version":"1.0","canonical_sha256":"55e71b2f695b7f5f71d6c67e69383b2a5d8cdf8f24eca49c3ff9ffccdb8365cc","source":{"kind":"arxiv","id":"2605.07557","version":4},"attestation_state":"computed","paper":{"title":"Beyond Distribution Estimation: Simplex Anchored Structural Inference Towards Universal Semi-Supervised Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Representation-level structural inference via SAGE enables effective learning in universal semi-supervised settings without distribution estimation.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bo Han, Hanyang Li, Jie Yu, Jun Ma, Yaxin Hou, Yuheng Jia","submitted_at":"2026-05-08T10:33:09Z","abstract_excerpt":"Semi-supervised learning faces significant challenges in realistic scenarios where labeled data is scarce and unlabeled data follows unknown, arbitrary distributions. We formalize this critical yet under-explored paradigm as Universal Semi-supervised Learning (UniSSL). Existing methods typically leverage unlabeled data via pseudo-labeling. However, they often rely on the idealized assumption of a uniform unlabeled data distribution or require sufficient labeled data to estimate it. In the UniSSL setting, such dependencies lead to numerous erroneous pseudo-labels, thereby triggering representat"},"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":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2605.07557","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-08T10:33:09Z","cross_cats_sorted":[],"title_canon_sha256":"30aa677c6006bb2b2b970d657ea13680454707082cd91db194e12af62e91444b","abstract_canon_sha256":"69c8d2632c3ee75aae9adc4d702f7ff11d7373c729060d2466bec24e19cd6dfb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:00:41.332764Z","signature_b64":"YkUoIo1pAapyxfgFWF3BqbS9BImNwxF4bJY1eCZ/u8MVI3/CEq+ZssVWf+Ng2XFFO5v5mCFTJgX3sPRsOik7Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"55e71b2f695b7f5f71d6c67e69383b2a5d8cdf8f24eca49c3ff9ffccdb8365cc","last_reissued_at":"2026-05-20T00:00:41.332049Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:00:41.332049Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond Distribution Estimation: Simplex Anchored Structural Inference Towards Universal Semi-Supervised Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Representation-level structural inference via SAGE enables effective learning in universal semi-supervised settings without distribution estimation.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bo Han, Hanyang Li, Jie Yu, Jun Ma, Yaxin Hou, Yuheng Jia","submitted_at":"2026-05-08T10:33:09Z","abstract_excerpt":"Semi-supervised learning faces significant challenges in realistic scenarios where labeled data is scarce and unlabeled data follows unknown, arbitrary distributions. We formalize this critical yet under-explored paradigm as Universal Semi-supervised Learning (UniSSL). Existing methods typically leverage unlabeled data via pseudo-labeling. However, they often rely on the idealized assumption of a uniform unlabeled data distribution or require sufficient labeled data to estimate it. In the UniSSL setting, such dependencies lead to numerous erroneous pseudo-labels, thereby triggering representat"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"By shifting focus to representation-level structural inference via Simplex Anchored Graph-state Equipartition (SAGE), the method captures high-order inter-sample dependencies to establish structural consensus, uses simplex equiangular tight frame vectors for inter-class separation, and achieves consistent outperformance with an average accuracy gain of 8.52% on five benchmarks.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The observation that inter-sample relations captured by representations are more reliable than pseudo-labels holds in the UniSSL setting with scarce labels and arbitrary unlabeled distributions; this premise is invoked to justify bypassing distribution estimation entirely.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"SAGE uses simplex-anchored graph-state equipartition and equiangular tight frames to perform structural inference on representations, bypassing distribution estimation in universal semi-supervised learning and achieving 8.52% average accuracy gains on benchmarks.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Representation-level structural inference via SAGE enables effective learning in universal semi-supervised settings without distribution estimation.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"8ef9bcf14bb01ff9b3ec0046ecc11ad63e57650d52657079d21001f1d7f9c466"},"source":{"id":"2605.07557","kind":"arxiv","version":4},"verdict":{"id":"5f022404-2837-4312-83f1-9f631e838c4a","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T16:34:57.638676Z","strongest_claim":"By shifting focus to representation-level structural inference via Simplex Anchored Graph-state Equipartition (SAGE), the method captures high-order inter-sample dependencies to establish structural consensus, uses simplex equiangular tight frame vectors for inter-class separation, and achieves consistent outperformance with an average accuracy gain of 8.52% on five benchmarks.","one_line_summary":"SAGE uses simplex-anchored graph-state equipartition and equiangular tight frames to perform structural inference on representations, bypassing distribution estimation in universal semi-supervised learning and achieving 8.52% average accuracy gains on benchmarks.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The observation that inter-sample relations captured by representations are more reliable than pseudo-labels holds in the UniSSL setting with scarce labels and arbitrary unlabeled distributions; this premise is invoked to justify bypassing distribution estimation entirely.","pith_extraction_headline":"Representation-level structural inference via SAGE enables effective learning in universal semi-supervised settings without distribution estimation."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.07557/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-19T16:31:18.571832Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T11:41:53.370025Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"7a72ea160c62f9a14d710712dfcc3c268deb61cb7b9808ad21160abb42926366"},"references":{"count":83,"sample":[{"doi":"","year":null,"title":"Medical Image Computing and Computer Assisted Intervention (MICCAI) , volume =","work_id":"43c521ba-0383-4350-a412-d2e4182de7ca","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Medical Image Computing and Computer Assisted Intervention (MICCAI) , volume =","work_id":"b31ad29d-b9aa-419f-8fa2-1c6034038202","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"International Conference on Machine Learning (ICML) , year =","work_id":"31a1a8ac-d64c-4923-90b1-027351fde620","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Yue Fan and Dengxin Dai and Anna Kukleva and Bernt Schiele , title =","work_id":"651845e2-2f11-4d92-ab2a-7ccb0a742571","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Proceedings of the AAAI Conference on Artificial Intelligence (AAAI) , volume =","work_id":"b6913d59-e705-4e7d-a467-97d102a35d7d","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":83,"snapshot_sha256":"6904b6db7f452c0e515cc10c91b9aa964f3337269c9e03d7698f2e0549912080","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"025e43b553f4e757477443efbb8e28d10feecaeb6ff0a6519dfbb33ed73c591c"},"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":"2605.07557","created_at":"2026-05-20T00:00:41.332145+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.07557v4","created_at":"2026-05-20T00:00:41.332145+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.07557","created_at":"2026-05-20T00:00:41.332145+00:00"},{"alias_kind":"pith_short_12","alias_value":"KXTRWL3JLN7V","created_at":"2026-05-20T00:00:41.332145+00:00"},{"alias_kind":"pith_short_16","alias_value":"KXTRWL3JLN7V64OW","created_at":"2026-05-20T00:00:41.332145+00:00"},{"alias_kind":"pith_short_8","alias_value":"KXTRWL3J","created_at":"2026-05-20T00:00:41.332145+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KXTRWL3JLN7V64OWYZ7GSOB3FJ","json":"https://pith.science/pith/KXTRWL3JLN7V64OWYZ7GSOB3FJ.json","graph_json":"https://pith.science/api/pith-number/KXTRWL3JLN7V64OWYZ7GSOB3FJ/graph.json","events_json":"https://pith.science/api/pith-number/KXTRWL3JLN7V64OWYZ7GSOB3FJ/events.json","paper":"https://pith.science/paper/KXTRWL3J"},"agent_actions":{"view_html":"https://pith.science/pith/KXTRWL3JLN7V64OWYZ7GSOB3FJ","download_json":"https://pith.science/pith/KXTRWL3JLN7V64OWYZ7GSOB3FJ.json","view_paper":"https://pith.science/paper/KXTRWL3J","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.07557&json=true","fetch_graph":"https://pith.science/api/pith-number/KXTRWL3JLN7V64OWYZ7GSOB3FJ/graph.json","fetch_events":"https://pith.science/api/pith-number/KXTRWL3JLN7V64OWYZ7GSOB3FJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KXTRWL3JLN7V64OWYZ7GSOB3FJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KXTRWL3JLN7V64OWYZ7GSOB3FJ/action/storage_attestation","attest_author":"https://pith.science/pith/KXTRWL3JLN7V64OWYZ7GSOB3FJ/action/author_attestation","sign_citation":"https://pith.science/pith/KXTRWL3JLN7V64OWYZ7GSOB3FJ/action/citation_signature","submit_replication":"https://pith.science/pith/KXTRWL3JLN7V64OWYZ7GSOB3FJ/action/replication_record"}},"created_at":"2026-05-20T00:00:41.332145+00:00","updated_at":"2026-05-20T00:00:41.332145+00:00"}