{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:ZUBCXTJFLJZPN24ERRBDC3YWZZ","short_pith_number":"pith:ZUBCXTJF","canonical_record":{"source":{"id":"2402.12530","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-19T20:40:48Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"a6424b9db8cecb69994698f964590bcecba31ec028d920b5c0ec624f3fb12f24","abstract_canon_sha256":"4980a5da7803f64b34fb1846338acb7626898b4701be5f05ea96bc0ac82627c3"},"schema_version":"1.0"},"canonical_sha256":"cd022bcd255a72f6eb848c42316f16ce5e002219f6aefbd6f73c2a2d79b10476","source":{"kind":"arxiv","id":"2402.12530","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.12530","created_at":"2026-07-05T07:47:12Z"},{"alias_kind":"arxiv_version","alias_value":"2402.12530v1","created_at":"2026-07-05T07:47:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.12530","created_at":"2026-07-05T07:47:12Z"},{"alias_kind":"pith_short_12","alias_value":"ZUBCXTJFLJZP","created_at":"2026-07-05T07:47:12Z"},{"alias_kind":"pith_short_16","alias_value":"ZUBCXTJFLJZPN24E","created_at":"2026-07-05T07:47:12Z"},{"alias_kind":"pith_short_8","alias_value":"ZUBCXTJF","created_at":"2026-07-05T07:47:12Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:ZUBCXTJFLJZPN24ERRBDC3YWZZ","target":"record","payload":{"canonical_record":{"source":{"id":"2402.12530","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-19T20:40:48Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"a6424b9db8cecb69994698f964590bcecba31ec028d920b5c0ec624f3fb12f24","abstract_canon_sha256":"4980a5da7803f64b34fb1846338acb7626898b4701be5f05ea96bc0ac82627c3"},"schema_version":"1.0"},"canonical_sha256":"cd022bcd255a72f6eb848c42316f16ce5e002219f6aefbd6f73c2a2d79b10476","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:47:12.654311Z","signature_b64":"WEg+/M54ULw0UPO5V8fKQbRehnmLk5QZll0z8dNJ/uE/bjiS+ca0av6Z1NDgI/k+UwC9kYP2Qkk1ioX6LcpgDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cd022bcd255a72f6eb848c42316f16ce5e002219f6aefbd6f73c2a2d79b10476","last_reissued_at":"2026-07-05T07:47:12.653902Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:47:12.653902Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2402.12530","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:47:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qi1ULHREESYblE6E5utOOySME85WPg2YNCLo3xE4wHlSxiDFKso3H0p0hW5x01w0vzUQqF84+Uksn91lMBdpBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T22:05:21.500386Z"},"content_sha256":"7aa7808628c547a90dd2d6dc0c17bb8caa214abbfad2022049c0ea60a01344ff","schema_version":"1.0","event_id":"sha256:7aa7808628c547a90dd2d6dc0c17bb8caa214abbfad2022049c0ea60a01344ff"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:ZUBCXTJFLJZPN24ERRBDC3YWZZ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Parallel Structures in Pre-training Data Yield In-Context Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Chen Zhao, He He, Kathleen McKeown, Yanda Chen, Zhou Yu","submitted_at":"2024-02-19T20:40:48Z","abstract_excerpt":"Pre-trained language models (LMs) are capable of in-context learning (ICL): they can adapt to a task with only a few examples given in the prompt without any parameter update. However, it is unclear where this capability comes from as there is a stark distribution shift between pre-training text and ICL prompts. In this work, we study what patterns of the pre-training data contribute to ICL. We find that LMs' ICL ability depends on $\\textit{parallel structures}$ in the pre-training data -- pairs of phrases following similar templates in the same context window. Specifically, we detect parallel"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.12530","kind":"arxiv","version":1},"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/2402.12530/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":"2026-07-05T07:47:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"y/3aPF/gaVzz+rzK2JkPXwfh3jAOA8fKwALZx1o7/yZ+LwPQE/ahS8IVgcbw3LzKu1sHTOu6vg99CYH56IYECQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T22:05:21.501018Z"},"content_sha256":"92787c7d829e06ba9f7f558f7fa0e572a17ca2460cad5811156e3d815f1025a7","schema_version":"1.0","event_id":"sha256:92787c7d829e06ba9f7f558f7fa0e572a17ca2460cad5811156e3d815f1025a7"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZUBCXTJFLJZPN24ERRBDC3YWZZ/bundle.json","state_url":"https://pith.science/pith/ZUBCXTJFLJZPN24ERRBDC3YWZZ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZUBCXTJFLJZPN24ERRBDC3YWZZ/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":"2026-08-17T22:05:21Z","links":{"resolver":"https://pith.science/pith/ZUBCXTJFLJZPN24ERRBDC3YWZZ","bundle":"https://pith.science/pith/ZUBCXTJFLJZPN24ERRBDC3YWZZ/bundle.json","state":"https://pith.science/pith/ZUBCXTJFLJZPN24ERRBDC3YWZZ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZUBCXTJFLJZPN24ERRBDC3YWZZ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:ZUBCXTJFLJZPN24ERRBDC3YWZZ","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":"4980a5da7803f64b34fb1846338acb7626898b4701be5f05ea96bc0ac82627c3","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-19T20:40:48Z","title_canon_sha256":"a6424b9db8cecb69994698f964590bcecba31ec028d920b5c0ec624f3fb12f24"},"schema_version":"1.0","source":{"id":"2402.12530","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.12530","created_at":"2026-07-05T07:47:12Z"},{"alias_kind":"arxiv_version","alias_value":"2402.12530v1","created_at":"2026-07-05T07:47:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.12530","created_at":"2026-07-05T07:47:12Z"},{"alias_kind":"pith_short_12","alias_value":"ZUBCXTJFLJZP","created_at":"2026-07-05T07:47:12Z"},{"alias_kind":"pith_short_16","alias_value":"ZUBCXTJFLJZPN24E","created_at":"2026-07-05T07:47:12Z"},{"alias_kind":"pith_short_8","alias_value":"ZUBCXTJF","created_at":"2026-07-05T07:47:12Z"}],"graph_snapshots":[{"event_id":"sha256:92787c7d829e06ba9f7f558f7fa0e572a17ca2460cad5811156e3d815f1025a7","target":"graph","created_at":"2026-07-05T07:47:12Z","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/2402.12530/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Pre-trained language models (LMs) are capable of in-context learning (ICL): they can adapt to a task with only a few examples given in the prompt without any parameter update. However, it is unclear where this capability comes from as there is a stark distribution shift between pre-training text and ICL prompts. In this work, we study what patterns of the pre-training data contribute to ICL. We find that LMs' ICL ability depends on $\\textit{parallel structures}$ in the pre-training data -- pairs of phrases following similar templates in the same context window. Specifically, we detect parallel","authors_text":"Chen Zhao, He He, Kathleen McKeown, Yanda Chen, Zhou Yu","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-19T20:40:48Z","title":"Parallel Structures in Pre-training Data Yield In-Context Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.12530","kind":"arxiv","version":1},"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":"sha256:7aa7808628c547a90dd2d6dc0c17bb8caa214abbfad2022049c0ea60a01344ff","target":"record","created_at":"2026-07-05T07:47:12Z","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":"4980a5da7803f64b34fb1846338acb7626898b4701be5f05ea96bc0ac82627c3","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-19T20:40:48Z","title_canon_sha256":"a6424b9db8cecb69994698f964590bcecba31ec028d920b5c0ec624f3fb12f24"},"schema_version":"1.0","source":{"id":"2402.12530","kind":"arxiv","version":1}},"canonical_sha256":"cd022bcd255a72f6eb848c42316f16ce5e002219f6aefbd6f73c2a2d79b10476","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cd022bcd255a72f6eb848c42316f16ce5e002219f6aefbd6f73c2a2d79b10476","first_computed_at":"2026-07-05T07:47:12.653902Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:47:12.653902Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"WEg+/M54ULw0UPO5V8fKQbRehnmLk5QZll0z8dNJ/uE/bjiS+ca0av6Z1NDgI/k+UwC9kYP2Qkk1ioX6LcpgDA==","signature_status":"signed_v1","signed_at":"2026-07-05T07:47:12.654311Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.12530","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7aa7808628c547a90dd2d6dc0c17bb8caa214abbfad2022049c0ea60a01344ff","sha256:92787c7d829e06ba9f7f558f7fa0e572a17ca2460cad5811156e3d815f1025a7"],"state_sha256":"bcee5a89033bb47db04e1fdcbca098681424f60931c17d935b4e0773ea135a95"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AeQizdRMtto07FGOYhrFN/+cCWCHmkFGBSTjxgXxKI7tcSL7u+02Hw2GG2THjt3Hi2cCxQUo3I8WGZwmP5naCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-17T22:05:21.505469Z","bundle_sha256":"d700623e70bb7dd0d28d17d6860cac9b971204e6158c5f01d3f96bf8810fcd10"}}