{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:LJSMJK2NKPP5LGLR2AEMCOZ6PT","short_pith_number":"pith:LJSMJK2N","canonical_record":{"source":{"id":"2010.03813","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-08T07:40:21Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"ae68b01def93249a5f5e37ec63b1a7a56fd43bffd978784fb129ff077a5fa85f","abstract_canon_sha256":"96f53579918cdcf762e8d3d6b74dba49288c9f9e5e5a3e5f69385d71f7b779b9"},"schema_version":"1.0"},"canonical_sha256":"5a64c4ab4d53dfd59971d008c13b3e7cec121d68ce8c4dc9dc9727fc858de705","source":{"kind":"arxiv","id":"2010.03813","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2010.03813","created_at":"2026-07-05T01:41:34Z"},{"alias_kind":"arxiv_version","alias_value":"2010.03813v2","created_at":"2026-07-05T01:41:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.03813","created_at":"2026-07-05T01:41:34Z"},{"alias_kind":"pith_short_12","alias_value":"LJSMJK2NKPP5","created_at":"2026-07-05T01:41:34Z"},{"alias_kind":"pith_short_16","alias_value":"LJSMJK2NKPP5LGLR","created_at":"2026-07-05T01:41:34Z"},{"alias_kind":"pith_short_8","alias_value":"LJSMJK2N","created_at":"2026-07-05T01:41:34Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:LJSMJK2NKPP5LGLR2AEMCOZ6PT","target":"record","payload":{"canonical_record":{"source":{"id":"2010.03813","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-08T07:40:21Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"ae68b01def93249a5f5e37ec63b1a7a56fd43bffd978784fb129ff077a5fa85f","abstract_canon_sha256":"96f53579918cdcf762e8d3d6b74dba49288c9f9e5e5a3e5f69385d71f7b779b9"},"schema_version":"1.0"},"canonical_sha256":"5a64c4ab4d53dfd59971d008c13b3e7cec121d68ce8c4dc9dc9727fc858de705","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:41:34.299173Z","signature_b64":"/R1LxIKuDIFnj+vIPtuhIDJVz286hZZ3RR2v7wF/vKdtNt7bHx/RZLoHrJRCNFo3yJuBwnX8TzAhN7aAN+jIDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5a64c4ab4d53dfd59971d008c13b3e7cec121d68ce8c4dc9dc9727fc858de705","last_reissued_at":"2026-07-05T01:41:34.298791Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:41:34.298791Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2010.03813","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":"2026-07-05T01:41:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Og8eZJCWrEeJn2vGSlOtiIptVCgHKfBINj1nSB3jFpPf/vuCIrXl3VeBURKGZrkpuOUUvGiBoxf5j9/hLDyBDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T00:15:31.455941Z"},"content_sha256":"dcbfecd18e4eeafb9661e1e1a2616dd2e91d6206813ad1b21788e218a908573c","schema_version":"1.0","event_id":"sha256:dcbfecd18e4eeafb9661e1e1a2616dd2e91d6206813ad1b21788e218a908573c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:LJSMJK2NKPP5LGLR2AEMCOZ6PT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"On the importance of pre-training data volume for compact language models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Fran\\c{c}ois Fleuret, Martin d'Hoffschmidt, Vincent Micheli","submitted_at":"2020-10-08T07:40:21Z","abstract_excerpt":"Recent advances in language modeling have led to computationally intensive and resource-demanding state-of-the-art models. In an effort towards sustainable practices, we study the impact of pre-training data volume on compact language models. Multiple BERT-based models are trained on gradually increasing amounts of French text. Through fine-tuning on the French Question Answering Dataset (FQuAD), we observe that well-performing models are obtained with as little as 100 MB of text. In addition, we show that past critically low amounts of pre-training data, an intermediate pre-training step on t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.03813","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/2010.03813/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":"2026-07-05T01:41:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vkwHnNpdtFDXVMHdaZQbf7/SA9r4ODredkKd9y+T4Jy5jciB3Tz5BEi17Vm1nEyBe1CevpHOkkp+tEbPPLWpAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T00:15:31.456855Z"},"content_sha256":"d9329afea7ac58a6f1b04b656c57699979abd3c9e0be1897c18b62ca41a2d341","schema_version":"1.0","event_id":"sha256:d9329afea7ac58a6f1b04b656c57699979abd3c9e0be1897c18b62ca41a2d341"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LJSMJK2NKPP5LGLR2AEMCOZ6PT/bundle.json","state_url":"https://pith.science/pith/LJSMJK2NKPP5LGLR2AEMCOZ6PT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LJSMJK2NKPP5LGLR2AEMCOZ6PT/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":"2026-08-05T00:15:31Z","links":{"resolver":"https://pith.science/pith/LJSMJK2NKPP5LGLR2AEMCOZ6PT","bundle":"https://pith.science/pith/LJSMJK2NKPP5LGLR2AEMCOZ6PT/bundle.json","state":"https://pith.science/pith/LJSMJK2NKPP5LGLR2AEMCOZ6PT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LJSMJK2NKPP5LGLR2AEMCOZ6PT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:LJSMJK2NKPP5LGLR2AEMCOZ6PT","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":"96f53579918cdcf762e8d3d6b74dba49288c9f9e5e5a3e5f69385d71f7b779b9","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-08T07:40:21Z","title_canon_sha256":"ae68b01def93249a5f5e37ec63b1a7a56fd43bffd978784fb129ff077a5fa85f"},"schema_version":"1.0","source":{"id":"2010.03813","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2010.03813","created_at":"2026-07-05T01:41:34Z"},{"alias_kind":"arxiv_version","alias_value":"2010.03813v2","created_at":"2026-07-05T01:41:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.03813","created_at":"2026-07-05T01:41:34Z"},{"alias_kind":"pith_short_12","alias_value":"LJSMJK2NKPP5","created_at":"2026-07-05T01:41:34Z"},{"alias_kind":"pith_short_16","alias_value":"LJSMJK2NKPP5LGLR","created_at":"2026-07-05T01:41:34Z"},{"alias_kind":"pith_short_8","alias_value":"LJSMJK2N","created_at":"2026-07-05T01:41:34Z"}],"graph_snapshots":[{"event_id":"sha256:d9329afea7ac58a6f1b04b656c57699979abd3c9e0be1897c18b62ca41a2d341","target":"graph","created_at":"2026-07-05T01:41:34Z","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/2010.03813/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent advances in language modeling have led to computationally intensive and resource-demanding state-of-the-art models. In an effort towards sustainable practices, we study the impact of pre-training data volume on compact language models. Multiple BERT-based models are trained on gradually increasing amounts of French text. Through fine-tuning on the French Question Answering Dataset (FQuAD), we observe that well-performing models are obtained with as little as 100 MB of text. In addition, we show that past critically low amounts of pre-training data, an intermediate pre-training step on t","authors_text":"Fran\\c{c}ois Fleuret, Martin d'Hoffschmidt, Vincent Micheli","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-08T07:40:21Z","title":"On the importance of pre-training data volume for compact language models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.03813","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:dcbfecd18e4eeafb9661e1e1a2616dd2e91d6206813ad1b21788e218a908573c","target":"record","created_at":"2026-07-05T01:41:34Z","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":"96f53579918cdcf762e8d3d6b74dba49288c9f9e5e5a3e5f69385d71f7b779b9","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-08T07:40:21Z","title_canon_sha256":"ae68b01def93249a5f5e37ec63b1a7a56fd43bffd978784fb129ff077a5fa85f"},"schema_version":"1.0","source":{"id":"2010.03813","kind":"arxiv","version":2}},"canonical_sha256":"5a64c4ab4d53dfd59971d008c13b3e7cec121d68ce8c4dc9dc9727fc858de705","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5a64c4ab4d53dfd59971d008c13b3e7cec121d68ce8c4dc9dc9727fc858de705","first_computed_at":"2026-07-05T01:41:34.298791Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T01:41:34.298791Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"/R1LxIKuDIFnj+vIPtuhIDJVz286hZZ3RR2v7wF/vKdtNt7bHx/RZLoHrJRCNFo3yJuBwnX8TzAhN7aAN+jIDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T01:41:34.299173Z","signed_message":"canonical_sha256_bytes"},"source_id":"2010.03813","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:dcbfecd18e4eeafb9661e1e1a2616dd2e91d6206813ad1b21788e218a908573c","sha256:d9329afea7ac58a6f1b04b656c57699979abd3c9e0be1897c18b62ca41a2d341"],"state_sha256":"b915911bc8423341f0d57b22d75b5fc311ddc2a59b82d249a56a1111a2241337"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KcSerciiWTZUMTGgcVGlmdszPN6wuF2ioPNTpFieH7BEP448WhhebpHJ7PFQqfF2D7GUM/qZ6+mYzYU/mmkrCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T00:15:31.463115Z","bundle_sha256":"6daab0321e498de15eebee7c6e080a67f91d824032e3e49e1ceb583a6a94f2ed"}}