{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:VLY6BVOL235DJ4WYAE27TXJU2O","short_pith_number":"pith:VLY6BVOL","canonical_record":{"source":{"id":"2304.11158","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-04-21T17:58:31Z","cross_cats_sorted":[],"title_canon_sha256":"0bb2d61bcc5250bbed0e1ef06a762b9e712b9b0d1c1becfb2d911813d11596ec","abstract_canon_sha256":"a0858a5efb7302c68c2775e4a0b49dacc511c50d2dbbc8e17b446488c89e89b8"},"schema_version":"1.0"},"canonical_sha256":"aaf1e0d5cbd6fa34f2d80135f9dd34d3bb8475227d4c9f4447ddafe972830a04","source":{"kind":"arxiv","id":"2304.11158","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2304.11158","created_at":"2026-07-05T06:16:18Z"},{"alias_kind":"arxiv_version","alias_value":"2304.11158v2","created_at":"2026-07-05T06:16:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2304.11158","created_at":"2026-07-05T06:16:18Z"},{"alias_kind":"pith_short_12","alias_value":"VLY6BVOL235D","created_at":"2026-07-05T06:16:18Z"},{"alias_kind":"pith_short_16","alias_value":"VLY6BVOL235DJ4WY","created_at":"2026-07-05T06:16:18Z"},{"alias_kind":"pith_short_8","alias_value":"VLY6BVOL","created_at":"2026-07-05T06:16:18Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:VLY6BVOL235DJ4WYAE27TXJU2O","target":"record","payload":{"canonical_record":{"source":{"id":"2304.11158","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-04-21T17:58:31Z","cross_cats_sorted":[],"title_canon_sha256":"0bb2d61bcc5250bbed0e1ef06a762b9e712b9b0d1c1becfb2d911813d11596ec","abstract_canon_sha256":"a0858a5efb7302c68c2775e4a0b49dacc511c50d2dbbc8e17b446488c89e89b8"},"schema_version":"1.0"},"canonical_sha256":"aaf1e0d5cbd6fa34f2d80135f9dd34d3bb8475227d4c9f4447ddafe972830a04","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:16:18.100422Z","signature_b64":"+F1RXvG5/Ss5Mch6weBsK6V/ITrLl/uPsqRnxLykawqU+xQuzJYfbwrcXu9plcUy8QL6OFRED7FYJbjyphNGBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"aaf1e0d5cbd6fa34f2d80135f9dd34d3bb8475227d4c9f4447ddafe972830a04","last_reissued_at":"2026-07-05T06:16:18.099920Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:16:18.099920Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2304.11158","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":"2026-07-05T06:16:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8DKxeOawMgN+tSsdHw2fa8xsVdbbsyLvYsVKjX02nxA85/dD/q1qcu7qk2OboLSurfUvLNgon5gGRNrDo7k5Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T08:47:59.277221Z"},"content_sha256":"588bc70cdc0e4ffdf27af86399b89c4e4fd88d1a6def5109900e5649db6aef13","schema_version":"1.0","event_id":"sha256:588bc70cdc0e4ffdf27af86399b89c4e4fd88d1a6def5109900e5649db6aef13"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:VLY6BVOL235DJ4WYAE27TXJU2O","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Emergent and Predictable Memorization in Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Edward Raff, Hailey Schoelkopf, Lintang Sutawika, Quentin Anthony, Shivanshu Purohit, Stella Biderman, USVSN Sai Prashanth","submitted_at":"2023-04-21T17:58:31Z","abstract_excerpt":"Memorization, or the tendency of large language models (LLMs) to output entire sequences from their training data verbatim, is a key concern for safely deploying language models. In particular, it is vital to minimize a model's memorization of sensitive datapoints such as those containing personal identifiable information (PII). The prevalence of such undesirable memorization can pose issues for model trainers, and may even require discarding an otherwise functional model. We therefore seek to predict which sequences will be memorized before a large model's full train-time by extrapolating the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2304.11158","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/2304.11158/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":"2026-07-05T06:16:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CoSN7MjY4AQmblTbOCYb8hofXoXb32S0Rk25f1kajhdNJFlUnZgWV6zdl/whfcGbk8OdTDFVJNbJg/R5bG3KCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T08:47:59.278150Z"},"content_sha256":"a4ef603a08ec7600547259fcfb23d456e6e87e5a52eba5d26f46d75d8fd30b4c","schema_version":"1.0","event_id":"sha256:a4ef603a08ec7600547259fcfb23d456e6e87e5a52eba5d26f46d75d8fd30b4c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/VLY6BVOL235DJ4WYAE27TXJU2O/bundle.json","state_url":"https://pith.science/pith/VLY6BVOL235DJ4WYAE27TXJU2O/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/VLY6BVOL235DJ4WYAE27TXJU2O/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":"2026-08-10T08:47:59Z","links":{"resolver":"https://pith.science/pith/VLY6BVOL235DJ4WYAE27TXJU2O","bundle":"https://pith.science/pith/VLY6BVOL235DJ4WYAE27TXJU2O/bundle.json","state":"https://pith.science/pith/VLY6BVOL235DJ4WYAE27TXJU2O/state.json","well_known_bundle":"https://pith.science/.well-known/pith/VLY6BVOL235DJ4WYAE27TXJU2O/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:VLY6BVOL235DJ4WYAE27TXJU2O","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":"a0858a5efb7302c68c2775e4a0b49dacc511c50d2dbbc8e17b446488c89e89b8","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-04-21T17:58:31Z","title_canon_sha256":"0bb2d61bcc5250bbed0e1ef06a762b9e712b9b0d1c1becfb2d911813d11596ec"},"schema_version":"1.0","source":{"id":"2304.11158","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2304.11158","created_at":"2026-07-05T06:16:18Z"},{"alias_kind":"arxiv_version","alias_value":"2304.11158v2","created_at":"2026-07-05T06:16:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2304.11158","created_at":"2026-07-05T06:16:18Z"},{"alias_kind":"pith_short_12","alias_value":"VLY6BVOL235D","created_at":"2026-07-05T06:16:18Z"},{"alias_kind":"pith_short_16","alias_value":"VLY6BVOL235DJ4WY","created_at":"2026-07-05T06:16:18Z"},{"alias_kind":"pith_short_8","alias_value":"VLY6BVOL","created_at":"2026-07-05T06:16:18Z"}],"graph_snapshots":[{"event_id":"sha256:a4ef603a08ec7600547259fcfb23d456e6e87e5a52eba5d26f46d75d8fd30b4c","target":"graph","created_at":"2026-07-05T06:16:18Z","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/2304.11158/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Memorization, or the tendency of large language models (LLMs) to output entire sequences from their training data verbatim, is a key concern for safely deploying language models. In particular, it is vital to minimize a model's memorization of sensitive datapoints such as those containing personal identifiable information (PII). The prevalence of such undesirable memorization can pose issues for model trainers, and may even require discarding an otherwise functional model. We therefore seek to predict which sequences will be memorized before a large model's full train-time by extrapolating the","authors_text":"Edward Raff, Hailey Schoelkopf, Lintang Sutawika, Quentin Anthony, Shivanshu Purohit, Stella Biderman, USVSN Sai Prashanth","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-04-21T17:58:31Z","title":"Emergent and Predictable Memorization in Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2304.11158","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":"sha256:588bc70cdc0e4ffdf27af86399b89c4e4fd88d1a6def5109900e5649db6aef13","target":"record","created_at":"2026-07-05T06:16:18Z","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":"a0858a5efb7302c68c2775e4a0b49dacc511c50d2dbbc8e17b446488c89e89b8","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-04-21T17:58:31Z","title_canon_sha256":"0bb2d61bcc5250bbed0e1ef06a762b9e712b9b0d1c1becfb2d911813d11596ec"},"schema_version":"1.0","source":{"id":"2304.11158","kind":"arxiv","version":2}},"canonical_sha256":"aaf1e0d5cbd6fa34f2d80135f9dd34d3bb8475227d4c9f4447ddafe972830a04","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"aaf1e0d5cbd6fa34f2d80135f9dd34d3bb8475227d4c9f4447ddafe972830a04","first_computed_at":"2026-07-05T06:16:18.099920Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:16:18.099920Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"+F1RXvG5/Ss5Mch6weBsK6V/ITrLl/uPsqRnxLykawqU+xQuzJYfbwrcXu9plcUy8QL6OFRED7FYJbjyphNGBg==","signature_status":"signed_v1","signed_at":"2026-07-05T06:16:18.100422Z","signed_message":"canonical_sha256_bytes"},"source_id":"2304.11158","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:588bc70cdc0e4ffdf27af86399b89c4e4fd88d1a6def5109900e5649db6aef13","sha256:a4ef603a08ec7600547259fcfb23d456e6e87e5a52eba5d26f46d75d8fd30b4c"],"state_sha256":"b17ed5c4aa2c3947b345f2dbd6ddb88acd46aad82b5b586101349154914bb761"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TWtb4NKEoZ/WQN8mYAUBXFBKfym+QLh/AYylfwlzCIxkFKlwia2Nbbkq79NDdeBTx86t8IX44XZQbF/WzFFCDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T08:47:59.284374Z","bundle_sha256":"5cada7555861a95479dc795694089560133a8281824bc3fedf1602a4650f4f15"}}