{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:BK4JYX4XAGVGJECY3WV3QOUBWM","short_pith_number":"pith:BK4JYX4X","canonical_record":{"source":{"id":"2010.02480","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-06T05:15:39Z","cross_cats_sorted":[],"title_canon_sha256":"93a6a925533c0b3e3a92abadfb0d5a37e83777866363bb139b959b3e09e3dbb6","abstract_canon_sha256":"2c80a723d884ff25bb67c85518dbe4b6686707c235ed1d3a35572e8712f3feb1"},"schema_version":"1.0"},"canonical_sha256":"0ab89c5f9701aa649058ddabb83a81b305440441e43dc969bfd1331d04cd4f86","source":{"kind":"arxiv","id":"2010.02480","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2010.02480","created_at":"2026-07-05T01:47:33Z"},{"alias_kind":"arxiv_version","alias_value":"2010.02480v2","created_at":"2026-07-05T01:47:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.02480","created_at":"2026-07-05T01:47:33Z"},{"alias_kind":"pith_short_12","alias_value":"BK4JYX4XAGVG","created_at":"2026-07-05T01:47:33Z"},{"alias_kind":"pith_short_16","alias_value":"BK4JYX4XAGVGJECY","created_at":"2026-07-05T01:47:33Z"},{"alias_kind":"pith_short_8","alias_value":"BK4JYX4X","created_at":"2026-07-05T01:47:33Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:BK4JYX4XAGVGJECY3WV3QOUBWM","target":"record","payload":{"canonical_record":{"source":{"id":"2010.02480","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-06T05:15:39Z","cross_cats_sorted":[],"title_canon_sha256":"93a6a925533c0b3e3a92abadfb0d5a37e83777866363bb139b959b3e09e3dbb6","abstract_canon_sha256":"2c80a723d884ff25bb67c85518dbe4b6686707c235ed1d3a35572e8712f3feb1"},"schema_version":"1.0"},"canonical_sha256":"0ab89c5f9701aa649058ddabb83a81b305440441e43dc969bfd1331d04cd4f86","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:47:33.003381Z","signature_b64":"KV+L0zWI2/YvJQaew7qVlRgpITKL3wd1ysrQ9UD+69IW0LHob2CPKLGIjnFkQfid5jGODCfVsyQ32+K8o6BiCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0ab89c5f9701aa649058ddabb83a81b305440441e43dc969bfd1331d04cd4f86","last_reissued_at":"2026-07-05T01:47:33.002910Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:47:33.002910Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2010.02480","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":"2026-07-05T01:47:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"T81yQDXCjxJ+wFiJ6FV0g0099rj/T4HjQSkd1MXOugTMXueT16uR5zstTtMG/UUY1c1hh/du1raFDjXDJOaVCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T03:10:01.718005Z"},"content_sha256":"1e29a5d378671c7800e7b5cfa21805d6805f03efc4e9cd853dc4c632fc45d533","schema_version":"1.0","event_id":"sha256:1e29a5d378671c7800e7b5cfa21805d6805f03efc4e9cd853dc4c632fc45d533"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:BK4JYX4XAGVGJECY3WV3QOUBWM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Pretrained Language Model Embryology: The Birth of ALBERT","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Cheng-Han Chiang, Hung-yi Lee, Sung-Feng Huang","submitted_at":"2020-10-06T05:15:39Z","abstract_excerpt":"While behaviors of pretrained language models (LMs) have been thoroughly examined, what happened during pretraining is rarely studied. We thus investigate the developmental process from a set of randomly initialized parameters to a totipotent language model, which we refer to as the embryology of a pretrained language model. Our results show that ALBERT learns to reconstruct and predict tokens of different parts of speech (POS) in different learning speeds during pretraining. We also find that linguistic knowledge and world knowledge do not generally improve as pretraining proceeds, nor do dow"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.02480","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/2010.02480/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":"2026-07-05T01:47:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SWRYnXuY0MxSZcAX+JMlAoMxblSwivZbG3/bIdUX1uLw7lvA91jtTHsk/pfPtTDXDtiT3MPj1pBHzbAo5xQ0Ag==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T03:10:01.718619Z"},"content_sha256":"586a2e3a415e556dd677742127c362e690ee7885867eb3eae2a5000f01e79bbe","schema_version":"1.0","event_id":"sha256:586a2e3a415e556dd677742127c362e690ee7885867eb3eae2a5000f01e79bbe"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BK4JYX4XAGVGJECY3WV3QOUBWM/bundle.json","state_url":"https://pith.science/pith/BK4JYX4XAGVGJECY3WV3QOUBWM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BK4JYX4XAGVGJECY3WV3QOUBWM/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":"2026-08-08T03:10:01Z","links":{"resolver":"https://pith.science/pith/BK4JYX4XAGVGJECY3WV3QOUBWM","bundle":"https://pith.science/pith/BK4JYX4XAGVGJECY3WV3QOUBWM/bundle.json","state":"https://pith.science/pith/BK4JYX4XAGVGJECY3WV3QOUBWM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BK4JYX4XAGVGJECY3WV3QOUBWM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:BK4JYX4XAGVGJECY3WV3QOUBWM","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":"2c80a723d884ff25bb67c85518dbe4b6686707c235ed1d3a35572e8712f3feb1","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-06T05:15:39Z","title_canon_sha256":"93a6a925533c0b3e3a92abadfb0d5a37e83777866363bb139b959b3e09e3dbb6"},"schema_version":"1.0","source":{"id":"2010.02480","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2010.02480","created_at":"2026-07-05T01:47:33Z"},{"alias_kind":"arxiv_version","alias_value":"2010.02480v2","created_at":"2026-07-05T01:47:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.02480","created_at":"2026-07-05T01:47:33Z"},{"alias_kind":"pith_short_12","alias_value":"BK4JYX4XAGVG","created_at":"2026-07-05T01:47:33Z"},{"alias_kind":"pith_short_16","alias_value":"BK4JYX4XAGVGJECY","created_at":"2026-07-05T01:47:33Z"},{"alias_kind":"pith_short_8","alias_value":"BK4JYX4X","created_at":"2026-07-05T01:47:33Z"}],"graph_snapshots":[{"event_id":"sha256:586a2e3a415e556dd677742127c362e690ee7885867eb3eae2a5000f01e79bbe","target":"graph","created_at":"2026-07-05T01:47:33Z","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/2010.02480/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"While behaviors of pretrained language models (LMs) have been thoroughly examined, what happened during pretraining is rarely studied. We thus investigate the developmental process from a set of randomly initialized parameters to a totipotent language model, which we refer to as the embryology of a pretrained language model. Our results show that ALBERT learns to reconstruct and predict tokens of different parts of speech (POS) in different learning speeds during pretraining. We also find that linguistic knowledge and world knowledge do not generally improve as pretraining proceeds, nor do dow","authors_text":"Cheng-Han Chiang, Hung-yi Lee, Sung-Feng Huang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-06T05:15:39Z","title":"Pretrained Language Model Embryology: The Birth of ALBERT"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.02480","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":"sha256:1e29a5d378671c7800e7b5cfa21805d6805f03efc4e9cd853dc4c632fc45d533","target":"record","created_at":"2026-07-05T01:47:33Z","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":"2c80a723d884ff25bb67c85518dbe4b6686707c235ed1d3a35572e8712f3feb1","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-06T05:15:39Z","title_canon_sha256":"93a6a925533c0b3e3a92abadfb0d5a37e83777866363bb139b959b3e09e3dbb6"},"schema_version":"1.0","source":{"id":"2010.02480","kind":"arxiv","version":2}},"canonical_sha256":"0ab89c5f9701aa649058ddabb83a81b305440441e43dc969bfd1331d04cd4f86","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0ab89c5f9701aa649058ddabb83a81b305440441e43dc969bfd1331d04cd4f86","first_computed_at":"2026-07-05T01:47:33.002910Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T01:47:33.002910Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"KV+L0zWI2/YvJQaew7qVlRgpITKL3wd1ysrQ9UD+69IW0LHob2CPKLGIjnFkQfid5jGODCfVsyQ32+K8o6BiCw==","signature_status":"signed_v1","signed_at":"2026-07-05T01:47:33.003381Z","signed_message":"canonical_sha256_bytes"},"source_id":"2010.02480","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1e29a5d378671c7800e7b5cfa21805d6805f03efc4e9cd853dc4c632fc45d533","sha256:586a2e3a415e556dd677742127c362e690ee7885867eb3eae2a5000f01e79bbe"],"state_sha256":"065e90db8b9c094ab4dd4c9c95f225ed76eed30737e24960637a9bbf26c87cd5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cg3IKcKMeBj2ntDScbhvr/kXT9SLNyf9xMGRyAm4sKMpNDrSXWNWzArlREgeH/EVSvStV/yM/b5eofEtYgSwDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T03:10:01.723141Z","bundle_sha256":"df3251bad5cab6de14e7e5284a3b6ffefa43200a191a3c2c65a8383a30b16af8"}}