{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:DVBXQN3HTOC66ZYODVSNLI36WI","short_pith_number":"pith:DVBXQN3H","canonical_record":{"source":{"id":"2210.10951","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-10-20T01:45:02Z","cross_cats_sorted":[],"title_canon_sha256":"819a79de0a73d5c92586960cb8a52d6e69b770320c13d1e169199e612b34dffd","abstract_canon_sha256":"d35d75bbd4128a08c0e6764670b36551beb50b423f86f1510ccc0913e25d189c"},"schema_version":"1.0"},"canonical_sha256":"1d437837679b85ef670e1d64d5a37eb202d9258ff8be32281d50158fa84f419e","source":{"kind":"arxiv","id":"2210.10951","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2210.10951","created_at":"2026-07-05T05:10:41Z"},{"alias_kind":"arxiv_version","alias_value":"2210.10951v2","created_at":"2026-07-05T05:10:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.10951","created_at":"2026-07-05T05:10:41Z"},{"alias_kind":"pith_short_12","alias_value":"DVBXQN3HTOC6","created_at":"2026-07-05T05:10:41Z"},{"alias_kind":"pith_short_16","alias_value":"DVBXQN3HTOC66ZYO","created_at":"2026-07-05T05:10:41Z"},{"alias_kind":"pith_short_8","alias_value":"DVBXQN3H","created_at":"2026-07-05T05:10:41Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:DVBXQN3HTOC66ZYODVSNLI36WI","target":"record","payload":{"canonical_record":{"source":{"id":"2210.10951","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-10-20T01:45:02Z","cross_cats_sorted":[],"title_canon_sha256":"819a79de0a73d5c92586960cb8a52d6e69b770320c13d1e169199e612b34dffd","abstract_canon_sha256":"d35d75bbd4128a08c0e6764670b36551beb50b423f86f1510ccc0913e25d189c"},"schema_version":"1.0"},"canonical_sha256":"1d437837679b85ef670e1d64d5a37eb202d9258ff8be32281d50158fa84f419e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:10:41.193716Z","signature_b64":"304R/ewzuqtHJKiqIwAI3oYiyUs8ctueap81+CuO08itIKNGbrkG1lN2p1JTN3eyrEq9GIma23izfbFxdypCAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1d437837679b85ef670e1d64d5a37eb202d9258ff8be32281d50158fa84f419e","last_reissued_at":"2026-07-05T05:10:41.193260Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:10:41.193260Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2210.10951","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T05:10:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"i90RHCH859tpxN6W5RJq9MePZXplf5ZObopdN0kBFbiDdwQ9hYWVgyh9u+IWXAekpS7lOMkqInNE+17YjnjMCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T18:49:17.229206Z"},"content_sha256":"da3dac2ee57e617a951f62cc024945fa2ce50142e718ef3a98a104c3b8b5dafa","schema_version":"1.0","event_id":"sha256:da3dac2ee57e617a951f62cc024945fa2ce50142e718ef3a98a104c3b8b5dafa"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:DVBXQN3HTOC66ZYODVSNLI36WI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Automatic Document Selection for Efficient Encoder Pretraining","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Benjamin Van Durme, Jo\\~ao Sedoc, Patrick Xia, Yukun Feng","submitted_at":"2022-10-20T01:45:02Z","abstract_excerpt":"Building pretrained language models is considered expensive and data-intensive, but must we increase dataset size to achieve better performance? We propose an alternative to larger training sets by automatically identifying smaller yet domain-representative subsets. We extend Cynical Data Selection, a statistical sentence scoring method that conditions on a representative target domain corpus. As an example, we treat the OntoNotes corpus as a target domain and pretrain a RoBERTa-like encoder from a cynically selected subset of the Pile. On both perplexity and across several downstream tasks in"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.10951","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/2210.10951/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T05:10:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Fck14yVNryBDKrtMGLPxb7OrC3DMEAwdQava18bvqm4ZoTKG4mBCU+2ugpnSryxypXWgVM+ThLMyL6MBu80wDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T18:49:17.230111Z"},"content_sha256":"6729afbc98f8436898b5f5f6fca4ec4e4cc06289309fd2e63dd7e351843580f2","schema_version":"1.0","event_id":"sha256:6729afbc98f8436898b5f5f6fca4ec4e4cc06289309fd2e63dd7e351843580f2"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/DVBXQN3HTOC66ZYODVSNLI36WI/bundle.json","state_url":"https://pith.science/pith/DVBXQN3HTOC66ZYODVSNLI36WI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/DVBXQN3HTOC66ZYODVSNLI36WI/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-07T18:49:17Z","links":{"resolver":"https://pith.science/pith/DVBXQN3HTOC66ZYODVSNLI36WI","bundle":"https://pith.science/pith/DVBXQN3HTOC66ZYODVSNLI36WI/bundle.json","state":"https://pith.science/pith/DVBXQN3HTOC66ZYODVSNLI36WI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/DVBXQN3HTOC66ZYODVSNLI36WI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:DVBXQN3HTOC66ZYODVSNLI36WI","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"d35d75bbd4128a08c0e6764670b36551beb50b423f86f1510ccc0913e25d189c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-10-20T01:45:02Z","title_canon_sha256":"819a79de0a73d5c92586960cb8a52d6e69b770320c13d1e169199e612b34dffd"},"schema_version":"1.0","source":{"id":"2210.10951","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2210.10951","created_at":"2026-07-05T05:10:41Z"},{"alias_kind":"arxiv_version","alias_value":"2210.10951v2","created_at":"2026-07-05T05:10:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.10951","created_at":"2026-07-05T05:10:41Z"},{"alias_kind":"pith_short_12","alias_value":"DVBXQN3HTOC6","created_at":"2026-07-05T05:10:41Z"},{"alias_kind":"pith_short_16","alias_value":"DVBXQN3HTOC66ZYO","created_at":"2026-07-05T05:10:41Z"},{"alias_kind":"pith_short_8","alias_value":"DVBXQN3H","created_at":"2026-07-05T05:10:41Z"}],"graph_snapshots":[{"event_id":"sha256:6729afbc98f8436898b5f5f6fca4ec4e4cc06289309fd2e63dd7e351843580f2","target":"graph","created_at":"2026-07-05T05:10:41Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2210.10951/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Building pretrained language models is considered expensive and data-intensive, but must we increase dataset size to achieve better performance? We propose an alternative to larger training sets by automatically identifying smaller yet domain-representative subsets. We extend Cynical Data Selection, a statistical sentence scoring method that conditions on a representative target domain corpus. As an example, we treat the OntoNotes corpus as a target domain and pretrain a RoBERTa-like encoder from a cynically selected subset of the Pile. On both perplexity and across several downstream tasks in","authors_text":"Benjamin Van Durme, Jo\\~ao Sedoc, Patrick Xia, Yukun Feng","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-10-20T01:45:02Z","title":"Automatic Document Selection for Efficient Encoder Pretraining"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.10951","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:da3dac2ee57e617a951f62cc024945fa2ce50142e718ef3a98a104c3b8b5dafa","target":"record","created_at":"2026-07-05T05:10:41Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"d35d75bbd4128a08c0e6764670b36551beb50b423f86f1510ccc0913e25d189c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-10-20T01:45:02Z","title_canon_sha256":"819a79de0a73d5c92586960cb8a52d6e69b770320c13d1e169199e612b34dffd"},"schema_version":"1.0","source":{"id":"2210.10951","kind":"arxiv","version":2}},"canonical_sha256":"1d437837679b85ef670e1d64d5a37eb202d9258ff8be32281d50158fa84f419e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1d437837679b85ef670e1d64d5a37eb202d9258ff8be32281d50158fa84f419e","first_computed_at":"2026-07-05T05:10:41.193260Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:10:41.193260Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"304R/ewzuqtHJKiqIwAI3oYiyUs8ctueap81+CuO08itIKNGbrkG1lN2p1JTN3eyrEq9GIma23izfbFxdypCAA==","signature_status":"signed_v1","signed_at":"2026-07-05T05:10:41.193716Z","signed_message":"canonical_sha256_bytes"},"source_id":"2210.10951","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:da3dac2ee57e617a951f62cc024945fa2ce50142e718ef3a98a104c3b8b5dafa","sha256:6729afbc98f8436898b5f5f6fca4ec4e4cc06289309fd2e63dd7e351843580f2"],"state_sha256":"718273ab152579f22fc7a02159dff4a3cd0338de414faef0ca02f2c4003e414b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3jxYQaI7rDUSx+GRiA3V8EJe7wUzoiypnL23ZbcO+ao0rJg4u1V8mCscFIlA3ZH3W8S/Ziguv9hoNuO/uT0sCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T18:49:17.237451Z","bundle_sha256":"1ffaada091f8631fc58d2f399dcd4c3bc52093e45cb65630ee14109b83fd699b"}}