{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:IB56YPAMHGFDCGTZDOTL5E7HGH","short_pith_number":"pith:IB56YPAM","canonical_record":{"source":{"id":"2409.04556","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-06T18:33:38Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"15accb5ba664c700c20d59bd846ec9a0551cf486295b91c0c1facba90d05205c","abstract_canon_sha256":"fd2d9f77184d75b6357cc765499343a54a7f2e0e76f219d6c93dcac338d75b5c"},"schema_version":"1.0"},"canonical_sha256":"407bec3c0c398a311a791ba6be93e731d45a6aa9c4d136c14229b0cb519330d2","source":{"kind":"arxiv","id":"2409.04556","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.04556","created_at":"2026-07-05T10:19:28Z"},{"alias_kind":"arxiv_version","alias_value":"2409.04556v2","created_at":"2026-07-05T10:19:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.04556","created_at":"2026-07-05T10:19:28Z"},{"alias_kind":"pith_short_12","alias_value":"IB56YPAMHGFD","created_at":"2026-07-05T10:19:28Z"},{"alias_kind":"pith_short_16","alias_value":"IB56YPAMHGFDCGTZ","created_at":"2026-07-05T10:19:28Z"},{"alias_kind":"pith_short_8","alias_value":"IB56YPAM","created_at":"2026-07-05T10:19:28Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:IB56YPAMHGFDCGTZDOTL5E7HGH","target":"record","payload":{"canonical_record":{"source":{"id":"2409.04556","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-06T18:33:38Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"15accb5ba664c700c20d59bd846ec9a0551cf486295b91c0c1facba90d05205c","abstract_canon_sha256":"fd2d9f77184d75b6357cc765499343a54a7f2e0e76f219d6c93dcac338d75b5c"},"schema_version":"1.0"},"canonical_sha256":"407bec3c0c398a311a791ba6be93e731d45a6aa9c4d136c14229b0cb519330d2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:19:28.018584Z","signature_b64":"OHuJoxw7FB+zlxyMWzJOITR6GgsdpuZfjxBvw3FiDbaRTaUomVO0LNwEBGwL6Ai+CEfrxRD/V6DwVFXGJdzHDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"407bec3c0c398a311a791ba6be93e731d45a6aa9c4d136c14229b0cb519330d2","last_reissued_at":"2026-07-05T10:19:28.018104Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:19:28.018104Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2409.04556","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":"2026-07-05T10:19:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PD6RKmDXiyLuT0OabRnk51LSM13M80mS8dsZ/ZHoj3RyFEPSDT0e+U3gEnX46BDijskoQS3ROX6LBUVpXOThBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T18:14:33.603304Z"},"content_sha256":"f8c018557342cd58e35bda3cb7bd325ea956700ab0f3f73ddcb726786bfdafe0","schema_version":"1.0","event_id":"sha256:f8c018557342cd58e35bda3cb7bd325ea956700ab0f3f73ddcb726786bfdafe0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:IB56YPAMHGFDCGTZDOTL5E7HGH","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"How Does Code Pretraining Affect Language Model Task Performance?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Jackson Petty, Sjoerd van Steenkiste, Tal Linzen","submitted_at":"2024-09-06T18:33:38Z","abstract_excerpt":"Large language models are increasingly trained on corpora containing both natural language and non-linguistic data like source code. Aside from aiding programming-related tasks, anecdotal evidence suggests that including code in pretraining corpora may improve performance on other, unrelated tasks, yet to date no work has been able to establish a causal connection by controlling between language and code data. Here we do just this. We pretrain language models on datasets which interleave natural language and code in two different settings: additive, in which the total volume of data seen durin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.04556","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/2409.04556/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":"2026-07-05T10:19:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"K+gq56+v3RiELgP6tNyt1elZtPk7QQxQVY9Lw7UCL+1K5Xmc6E4w76d9wVxo95PzwhTo+KHbnW8HobOIEeNEDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T18:14:33.604210Z"},"content_sha256":"ea7d36f63b07066545e06e5bfbf33a84c0cd55698983a311ede46acb69c8acef","schema_version":"1.0","event_id":"sha256:ea7d36f63b07066545e06e5bfbf33a84c0cd55698983a311ede46acb69c8acef"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/IB56YPAMHGFDCGTZDOTL5E7HGH/bundle.json","state_url":"https://pith.science/pith/IB56YPAMHGFDCGTZDOTL5E7HGH/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/IB56YPAMHGFDCGTZDOTL5E7HGH/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":"2026-08-03T18:14:33Z","links":{"resolver":"https://pith.science/pith/IB56YPAMHGFDCGTZDOTL5E7HGH","bundle":"https://pith.science/pith/IB56YPAMHGFDCGTZDOTL5E7HGH/bundle.json","state":"https://pith.science/pith/IB56YPAMHGFDCGTZDOTL5E7HGH/state.json","well_known_bundle":"https://pith.science/.well-known/pith/IB56YPAMHGFDCGTZDOTL5E7HGH/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:IB56YPAMHGFDCGTZDOTL5E7HGH","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":"fd2d9f77184d75b6357cc765499343a54a7f2e0e76f219d6c93dcac338d75b5c","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-06T18:33:38Z","title_canon_sha256":"15accb5ba664c700c20d59bd846ec9a0551cf486295b91c0c1facba90d05205c"},"schema_version":"1.0","source":{"id":"2409.04556","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.04556","created_at":"2026-07-05T10:19:28Z"},{"alias_kind":"arxiv_version","alias_value":"2409.04556v2","created_at":"2026-07-05T10:19:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.04556","created_at":"2026-07-05T10:19:28Z"},{"alias_kind":"pith_short_12","alias_value":"IB56YPAMHGFD","created_at":"2026-07-05T10:19:28Z"},{"alias_kind":"pith_short_16","alias_value":"IB56YPAMHGFDCGTZ","created_at":"2026-07-05T10:19:28Z"},{"alias_kind":"pith_short_8","alias_value":"IB56YPAM","created_at":"2026-07-05T10:19:28Z"}],"graph_snapshots":[{"event_id":"sha256:ea7d36f63b07066545e06e5bfbf33a84c0cd55698983a311ede46acb69c8acef","target":"graph","created_at":"2026-07-05T10:19:28Z","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/2409.04556/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models are increasingly trained on corpora containing both natural language and non-linguistic data like source code. Aside from aiding programming-related tasks, anecdotal evidence suggests that including code in pretraining corpora may improve performance on other, unrelated tasks, yet to date no work has been able to establish a causal connection by controlling between language and code data. Here we do just this. We pretrain language models on datasets which interleave natural language and code in two different settings: additive, in which the total volume of data seen durin","authors_text":"Jackson Petty, Sjoerd van Steenkiste, Tal Linzen","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-06T18:33:38Z","title":"How Does Code Pretraining Affect Language Model Task Performance?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.04556","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:f8c018557342cd58e35bda3cb7bd325ea956700ab0f3f73ddcb726786bfdafe0","target":"record","created_at":"2026-07-05T10:19:28Z","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":"fd2d9f77184d75b6357cc765499343a54a7f2e0e76f219d6c93dcac338d75b5c","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-06T18:33:38Z","title_canon_sha256":"15accb5ba664c700c20d59bd846ec9a0551cf486295b91c0c1facba90d05205c"},"schema_version":"1.0","source":{"id":"2409.04556","kind":"arxiv","version":2}},"canonical_sha256":"407bec3c0c398a311a791ba6be93e731d45a6aa9c4d136c14229b0cb519330d2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"407bec3c0c398a311a791ba6be93e731d45a6aa9c4d136c14229b0cb519330d2","first_computed_at":"2026-07-05T10:19:28.018104Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:19:28.018104Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"OHuJoxw7FB+zlxyMWzJOITR6GgsdpuZfjxBvw3FiDbaRTaUomVO0LNwEBGwL6Ai+CEfrxRD/V6DwVFXGJdzHDw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:19:28.018584Z","signed_message":"canonical_sha256_bytes"},"source_id":"2409.04556","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f8c018557342cd58e35bda3cb7bd325ea956700ab0f3f73ddcb726786bfdafe0","sha256:ea7d36f63b07066545e06e5bfbf33a84c0cd55698983a311ede46acb69c8acef"],"state_sha256":"d8ca5a0cfcd302c2f4af7fecdcc9abdb36f91d00e32b7c1accbf7a615554174a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"q+k2YmlMOaRdGvs2J2lR8ppB/VygJYRRiQcJkmPVmaTNGcq98qX7F7d8eK/hgrcs+qINQJVuKqZjyX+dVIcVBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T18:14:33.610895Z","bundle_sha256":"3286fe33537f09eedc54f907c1aff62b9e95b4463ff8cff03916fe0f6a0b393f"}}