{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:ORLEI7VKDCZH2APZ4JLHUWLLNI","short_pith_number":"pith:ORLEI7VK","schema_version":"1.0","canonical_sha256":"7456447eaa18b27d01f9e2567a596b6a20d955f5cc68c3ef4b0aded2f04fe9c5","source":{"kind":"arxiv","id":"2107.01057","version":1},"attestation_state":"computed","paper":{"title":"Backward-Compatible Prediction Updates: A Probabilistic Approach","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Bernhard Sch\\\"olkopf, Francesco Locatello, Frederik Tr\\\"auble, Julius von K\\\"ugelgen, Matth\\\"aus Kleindessner, Peter Gehler","submitted_at":"2021-07-02T13:05:31Z","abstract_excerpt":"When machine learning systems meet real world applications, accuracy is only one of several requirements. In this paper, we assay a complementary perspective originating from the increasing availability of pre-trained and regularly improving state-of-the-art models. While new improved models develop at a fast pace, downstream tasks vary more slowly or stay constant. Assume that we have a large unlabelled data set for which we want to maintain accurate predictions. Whenever a new and presumably better ML models becomes available, we encounter two problems: (i) given a limited budget, which data"},"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":"2107.01057","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-07-02T13:05:31Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"4e46b9c2be580eaf6a1737db3009d492a5638f94cd4324c49ee07d09cc58da6e","abstract_canon_sha256":"65da2e04e026e1a6a7195b612850ac35485bdddc1d6b163d7609ac824e866304"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:54:36.687495Z","signature_b64":"XqU8MYe1yakQ2SLBrOFiS3p4v2GLqUVvqCfFTHdg7EvzEc1hQ/ULZxyjhnzeFc3mJUO7NLA4L5zdOG4IJ6J1AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7456447eaa18b27d01f9e2567a596b6a20d955f5cc68c3ef4b0aded2f04fe9c5","last_reissued_at":"2026-07-05T02:54:36.687083Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:54:36.687083Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Backward-Compatible Prediction Updates: A Probabilistic Approach","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Bernhard Sch\\\"olkopf, Francesco Locatello, Frederik Tr\\\"auble, Julius von K\\\"ugelgen, Matth\\\"aus Kleindessner, Peter Gehler","submitted_at":"2021-07-02T13:05:31Z","abstract_excerpt":"When machine learning systems meet real world applications, accuracy is only one of several requirements. In this paper, we assay a complementary perspective originating from the increasing availability of pre-trained and regularly improving state-of-the-art models. While new improved models develop at a fast pace, downstream tasks vary more slowly or stay constant. Assume that we have a large unlabelled data set for which we want to maintain accurate predictions. Whenever a new and presumably better ML models becomes available, we encounter two problems: (i) given a limited budget, which data"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.01057","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/2107.01057/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":"2107.01057","created_at":"2026-07-05T02:54:36.687138+00:00"},{"alias_kind":"arxiv_version","alias_value":"2107.01057v1","created_at":"2026-07-05T02:54:36.687138+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.01057","created_at":"2026-07-05T02:54:36.687138+00:00"},{"alias_kind":"pith_short_12","alias_value":"ORLEI7VKDCZH","created_at":"2026-07-05T02:54:36.687138+00:00"},{"alias_kind":"pith_short_16","alias_value":"ORLEI7VKDCZH2APZ","created_at":"2026-07-05T02:54:36.687138+00:00"},{"alias_kind":"pith_short_8","alias_value":"ORLEI7VK","created_at":"2026-07-05T02:54:36.687138+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ORLEI7VKDCZH2APZ4JLHUWLLNI","json":"https://pith.science/pith/ORLEI7VKDCZH2APZ4JLHUWLLNI.json","graph_json":"https://pith.science/api/pith-number/ORLEI7VKDCZH2APZ4JLHUWLLNI/graph.json","events_json":"https://pith.science/api/pith-number/ORLEI7VKDCZH2APZ4JLHUWLLNI/events.json","paper":"https://pith.science/paper/ORLEI7VK"},"agent_actions":{"view_html":"https://pith.science/pith/ORLEI7VKDCZH2APZ4JLHUWLLNI","download_json":"https://pith.science/pith/ORLEI7VKDCZH2APZ4JLHUWLLNI.json","view_paper":"https://pith.science/paper/ORLEI7VK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2107.01057&json=true","fetch_graph":"https://pith.science/api/pith-number/ORLEI7VKDCZH2APZ4JLHUWLLNI/graph.json","fetch_events":"https://pith.science/api/pith-number/ORLEI7VKDCZH2APZ4JLHUWLLNI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ORLEI7VKDCZH2APZ4JLHUWLLNI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ORLEI7VKDCZH2APZ4JLHUWLLNI/action/storage_attestation","attest_author":"https://pith.science/pith/ORLEI7VKDCZH2APZ4JLHUWLLNI/action/author_attestation","sign_citation":"https://pith.science/pith/ORLEI7VKDCZH2APZ4JLHUWLLNI/action/citation_signature","submit_replication":"https://pith.science/pith/ORLEI7VKDCZH2APZ4JLHUWLLNI/action/replication_record"}},"created_at":"2026-07-05T02:54:36.687138+00:00","updated_at":"2026-07-05T02:54:36.687138+00:00"}