{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:LJPVDZY5GXNNSAXZWKRS4XB66M","short_pith_number":"pith:LJPVDZY5","canonical_record":{"source":{"id":"2502.06042","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-09T21:44:27Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"5428bf2b0b98b9a19eadc453b2e808807f2ea803fbaa7460547a82257f095812","abstract_canon_sha256":"50c96e441c5c4e7b50bfd37cbddb833fa30174883561c29add0b52ef221ae119"},"schema_version":"1.0"},"canonical_sha256":"5a5f51e71d35dad902f9b2a32e5c3ef32dbea4051da57779aec27daf76ed87f6","source":{"kind":"arxiv","id":"2502.06042","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.06042","created_at":"2026-07-05T11:10:04Z"},{"alias_kind":"arxiv_version","alias_value":"2502.06042v2","created_at":"2026-07-05T11:10:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.06042","created_at":"2026-07-05T11:10:04Z"},{"alias_kind":"pith_short_12","alias_value":"LJPVDZY5GXNN","created_at":"2026-07-05T11:10:04Z"},{"alias_kind":"pith_short_16","alias_value":"LJPVDZY5GXNNSAXZ","created_at":"2026-07-05T11:10:04Z"},{"alias_kind":"pith_short_8","alias_value":"LJPVDZY5","created_at":"2026-07-05T11:10:04Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:LJPVDZY5GXNNSAXZWKRS4XB66M","target":"record","payload":{"canonical_record":{"source":{"id":"2502.06042","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-09T21:44:27Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"5428bf2b0b98b9a19eadc453b2e808807f2ea803fbaa7460547a82257f095812","abstract_canon_sha256":"50c96e441c5c4e7b50bfd37cbddb833fa30174883561c29add0b52ef221ae119"},"schema_version":"1.0"},"canonical_sha256":"5a5f51e71d35dad902f9b2a32e5c3ef32dbea4051da57779aec27daf76ed87f6","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:10:04.604191Z","signature_b64":"igN7oZ0ORm6wQrFEZKsdWaQlG1mKnEFWC3EL8CZDNwK9QKLj/ZAjrVxe9mD2PgWGjReWxzpQF2DRqdDom4YrAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5a5f51e71d35dad902f9b2a32e5c3ef32dbea4051da57779aec27daf76ed87f6","last_reissued_at":"2026-07-05T11:10:04.603666Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:10:04.603666Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.06042","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":"2026-07-05T11:10:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HJczD+wuVA2T/mRvKLe6SOtYkkezuM6I7XPkAe9RGqu3U5OcDSH+SjVaZnxwd1ggXADe/V5f3tF7G3lyGbGVDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T02:29:42.399725Z"},"content_sha256":"d75f809c35d77fd4cfcd772d7256fd6644f4a8307574bb952f494e5d2f1cb326","schema_version":"1.0","event_id":"sha256:d75f809c35d77fd4cfcd772d7256fd6644f4a8307574bb952f494e5d2f1cb326"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:LJPVDZY5GXNNSAXZWKRS4XB66M","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Scaling Laws for Forgetting during Finetuning with Pretraining Data Injection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Dan Busbridge, David Grangier, Eleonora Gualdoni, Louis Bethune, Marco Cuturi, Pierre Ablin","submitted_at":"2025-02-09T21:44:27Z","abstract_excerpt":"A widespread strategy to obtain a language model that performs well on a target domain is to finetune a pretrained model to perform unsupervised next-token prediction on data from that target domain. Finetuning presents two challenges: (i) if the amount of target data is limited, as in most practical applications, the model will quickly overfit, and (ii) the model will drift away from the original model, forgetting the pretraining data and the generic knowledge that comes with it. We aim to derive scaling laws that quantify these two phenomena for various target domains, amounts of available t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.06042","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/2502.06042/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":"2026-07-05T11:10:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"oFXwtxEs5c0c3uUKXAVC06TzmnTtxTartDZ+1xhL4yBDDHGZrWugejN39j/JU12JZWNReIC8jd8Y+q2wGq/WCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T02:29:42.400255Z"},"content_sha256":"885cddc9d4532fd551a932dc1e236d25b0041f9cd5cdc50fdf03aed83438f4c4","schema_version":"1.0","event_id":"sha256:885cddc9d4532fd551a932dc1e236d25b0041f9cd5cdc50fdf03aed83438f4c4"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LJPVDZY5GXNNSAXZWKRS4XB66M/bundle.json","state_url":"https://pith.science/pith/LJPVDZY5GXNNSAXZWKRS4XB66M/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LJPVDZY5GXNNSAXZWKRS4XB66M/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":"2026-08-10T02:29:42Z","links":{"resolver":"https://pith.science/pith/LJPVDZY5GXNNSAXZWKRS4XB66M","bundle":"https://pith.science/pith/LJPVDZY5GXNNSAXZWKRS4XB66M/bundle.json","state":"https://pith.science/pith/LJPVDZY5GXNNSAXZWKRS4XB66M/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LJPVDZY5GXNNSAXZWKRS4XB66M/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:LJPVDZY5GXNNSAXZWKRS4XB66M","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":"50c96e441c5c4e7b50bfd37cbddb833fa30174883561c29add0b52ef221ae119","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-09T21:44:27Z","title_canon_sha256":"5428bf2b0b98b9a19eadc453b2e808807f2ea803fbaa7460547a82257f095812"},"schema_version":"1.0","source":{"id":"2502.06042","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.06042","created_at":"2026-07-05T11:10:04Z"},{"alias_kind":"arxiv_version","alias_value":"2502.06042v2","created_at":"2026-07-05T11:10:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.06042","created_at":"2026-07-05T11:10:04Z"},{"alias_kind":"pith_short_12","alias_value":"LJPVDZY5GXNN","created_at":"2026-07-05T11:10:04Z"},{"alias_kind":"pith_short_16","alias_value":"LJPVDZY5GXNNSAXZ","created_at":"2026-07-05T11:10:04Z"},{"alias_kind":"pith_short_8","alias_value":"LJPVDZY5","created_at":"2026-07-05T11:10:04Z"}],"graph_snapshots":[{"event_id":"sha256:885cddc9d4532fd551a932dc1e236d25b0041f9cd5cdc50fdf03aed83438f4c4","target":"graph","created_at":"2026-07-05T11:10:04Z","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/2502.06042/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"A widespread strategy to obtain a language model that performs well on a target domain is to finetune a pretrained model to perform unsupervised next-token prediction on data from that target domain. Finetuning presents two challenges: (i) if the amount of target data is limited, as in most practical applications, the model will quickly overfit, and (ii) the model will drift away from the original model, forgetting the pretraining data and the generic knowledge that comes with it. We aim to derive scaling laws that quantify these two phenomena for various target domains, amounts of available t","authors_text":"Dan Busbridge, David Grangier, Eleonora Gualdoni, Louis Bethune, Marco Cuturi, Pierre Ablin","cross_cats":["cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-09T21:44:27Z","title":"Scaling Laws for Forgetting during Finetuning with Pretraining Data Injection"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.06042","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:d75f809c35d77fd4cfcd772d7256fd6644f4a8307574bb952f494e5d2f1cb326","target":"record","created_at":"2026-07-05T11:10:04Z","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":"50c96e441c5c4e7b50bfd37cbddb833fa30174883561c29add0b52ef221ae119","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-09T21:44:27Z","title_canon_sha256":"5428bf2b0b98b9a19eadc453b2e808807f2ea803fbaa7460547a82257f095812"},"schema_version":"1.0","source":{"id":"2502.06042","kind":"arxiv","version":2}},"canonical_sha256":"5a5f51e71d35dad902f9b2a32e5c3ef32dbea4051da57779aec27daf76ed87f6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5a5f51e71d35dad902f9b2a32e5c3ef32dbea4051da57779aec27daf76ed87f6","first_computed_at":"2026-07-05T11:10:04.603666Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:10:04.603666Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"igN7oZ0ORm6wQrFEZKsdWaQlG1mKnEFWC3EL8CZDNwK9QKLj/ZAjrVxe9mD2PgWGjReWxzpQF2DRqdDom4YrAg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:10:04.604191Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.06042","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d75f809c35d77fd4cfcd772d7256fd6644f4a8307574bb952f494e5d2f1cb326","sha256:885cddc9d4532fd551a932dc1e236d25b0041f9cd5cdc50fdf03aed83438f4c4"],"state_sha256":"367decd6a60ed5548bdcd0f3a405539f08c46123e2775e6aaf4e2f3b60bbb951"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Uia6RXN7PB05jccMzsZYMDf2iR9iCQBrI16DUTtl4l3IwYL1wQ95NgC+Nk9gvPm46NF2NQFqSBfkA8mtr8pSCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T02:29:42.404414Z","bundle_sha256":"db5cfbffb774b0c09908c5581a30bb5597b7070ae36cc4c4fcadde671ab36055"}}