{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:KZU3EDUEQT4DMVI3MTLKSZZSMS","short_pith_number":"pith:KZU3EDUE","canonical_record":{"source":{"id":"2502.16802","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-24T03:25:56Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3d8c228d5434fbd556c0056d8bf7457bd7923f893b7fafa998a0be0e063466e7","abstract_canon_sha256":"da413d1ada2601b78f8abf1b2e65d30ff929bc429ec0ea51c899e2da09159646"},"schema_version":"1.0"},"canonical_sha256":"5669b20e8484f836551b64d6a9673264ab89584f6c4e5c5f1d6408f886de1e28","source":{"kind":"arxiv","id":"2502.16802","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.16802","created_at":"2026-07-05T11:50:31Z"},{"alias_kind":"arxiv_version","alias_value":"2502.16802v3","created_at":"2026-07-05T11:50:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.16802","created_at":"2026-07-05T11:50:31Z"},{"alias_kind":"pith_short_12","alias_value":"KZU3EDUEQT4D","created_at":"2026-07-05T11:50:31Z"},{"alias_kind":"pith_short_16","alias_value":"KZU3EDUEQT4DMVI3","created_at":"2026-07-05T11:50:31Z"},{"alias_kind":"pith_short_8","alias_value":"KZU3EDUE","created_at":"2026-07-05T11:50:31Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:KZU3EDUEQT4DMVI3MTLKSZZSMS","target":"record","payload":{"canonical_record":{"source":{"id":"2502.16802","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-24T03:25:56Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3d8c228d5434fbd556c0056d8bf7457bd7923f893b7fafa998a0be0e063466e7","abstract_canon_sha256":"da413d1ada2601b78f8abf1b2e65d30ff929bc429ec0ea51c899e2da09159646"},"schema_version":"1.0"},"canonical_sha256":"5669b20e8484f836551b64d6a9673264ab89584f6c4e5c5f1d6408f886de1e28","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:50:31.684095Z","signature_b64":"IdjvVkMFnafoOcKVyrNLKefFBy7v2nuU91D3f5DYNmIDX5BFjXY9dZr0bFIrLLtXXgmiBPcyqQArkgbDL97gAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5669b20e8484f836551b64d6a9673264ab89584f6c4e5c5f1d6408f886de1e28","last_reissued_at":"2026-07-05T11:50:31.683577Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:50:31.683577Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.16802","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:50:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MQI6rsSth7N0hMs3lstDwgNMTKoouXcBxda308lFyB6g3qut+LGh1+cbHAYox36iOnkmiqlQgQ8PayvCxw2wCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T22:09:48.840222Z"},"content_sha256":"b6d019ee108fd11f3c194a9c021c3f4d77280d20ea71b51919d685f4b4042a37","schema_version":"1.0","event_id":"sha256:b6d019ee108fd11f3c194a9c021c3f4d77280d20ea71b51919d685f4b4042a37"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:KZU3EDUEQT4DMVI3MTLKSZZSMS","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Topic Over Source: The Key to Effective Data Mixing for Language Models Pre-training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Conghui He, He Zhu, Jiahui Peng, Jiantao Qiu, Jing Yu, Ren Ma, Xinlin Zhuang","submitted_at":"2025-02-24T03:25:56Z","abstract_excerpt":"The performance of large language models (LLMs) is significantly affected by the quality and composition of their pre-training data, which is inherently diverse, spanning various languages, sources, and topics. Effectively integrating these heterogeneous data groups is crucial for optimizing LLM performance. Previous research has predominantly concentrated on source-based data mixing, often neglecting the nuanced topic-level characteristics of the data. To address this gap, we propose a topic-based data mixing strategy that utilizes detailed topic labels generated through a multi-stage process"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.16802","kind":"arxiv","version":3},"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/2502.16802/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":"2026-07-05T11:50:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"q1V69ecQ3z4492gqtJIDxaHUVnMaett9ci9KT2nmVYK3zp9D+hkWs6TjfuWNpyrgS1I9JeeCbj4ruoDacJHYCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T22:09:48.840784Z"},"content_sha256":"71dca2c4575d0f495cf5604ae69f59d1d5040e72c378d147c7d84f109727b3e5","schema_version":"1.0","event_id":"sha256:71dca2c4575d0f495cf5604ae69f59d1d5040e72c378d147c7d84f109727b3e5"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KZU3EDUEQT4DMVI3MTLKSZZSMS/bundle.json","state_url":"https://pith.science/pith/KZU3EDUEQT4DMVI3MTLKSZZSMS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KZU3EDUEQT4DMVI3MTLKSZZSMS/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":"2026-08-18T22:09:48Z","links":{"resolver":"https://pith.science/pith/KZU3EDUEQT4DMVI3MTLKSZZSMS","bundle":"https://pith.science/pith/KZU3EDUEQT4DMVI3MTLKSZZSMS/bundle.json","state":"https://pith.science/pith/KZU3EDUEQT4DMVI3MTLKSZZSMS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KZU3EDUEQT4DMVI3MTLKSZZSMS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:KZU3EDUEQT4DMVI3MTLKSZZSMS","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":"da413d1ada2601b78f8abf1b2e65d30ff929bc429ec0ea51c899e2da09159646","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-24T03:25:56Z","title_canon_sha256":"3d8c228d5434fbd556c0056d8bf7457bd7923f893b7fafa998a0be0e063466e7"},"schema_version":"1.0","source":{"id":"2502.16802","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.16802","created_at":"2026-07-05T11:50:31Z"},{"alias_kind":"arxiv_version","alias_value":"2502.16802v3","created_at":"2026-07-05T11:50:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.16802","created_at":"2026-07-05T11:50:31Z"},{"alias_kind":"pith_short_12","alias_value":"KZU3EDUEQT4D","created_at":"2026-07-05T11:50:31Z"},{"alias_kind":"pith_short_16","alias_value":"KZU3EDUEQT4DMVI3","created_at":"2026-07-05T11:50:31Z"},{"alias_kind":"pith_short_8","alias_value":"KZU3EDUE","created_at":"2026-07-05T11:50:31Z"}],"graph_snapshots":[{"event_id":"sha256:71dca2c4575d0f495cf5604ae69f59d1d5040e72c378d147c7d84f109727b3e5","target":"graph","created_at":"2026-07-05T11:50:31Z","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/2502.16802/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The performance of large language models (LLMs) is significantly affected by the quality and composition of their pre-training data, which is inherently diverse, spanning various languages, sources, and topics. Effectively integrating these heterogeneous data groups is crucial for optimizing LLM performance. Previous research has predominantly concentrated on source-based data mixing, often neglecting the nuanced topic-level characteristics of the data. To address this gap, we propose a topic-based data mixing strategy that utilizes detailed topic labels generated through a multi-stage process","authors_text":"Conghui He, He Zhu, Jiahui Peng, Jiantao Qiu, Jing Yu, Ren Ma, Xinlin Zhuang","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-24T03:25:56Z","title":"Topic Over Source: The Key to Effective Data Mixing for Language Models Pre-training"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.16802","kind":"arxiv","version":3},"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:b6d019ee108fd11f3c194a9c021c3f4d77280d20ea71b51919d685f4b4042a37","target":"record","created_at":"2026-07-05T11:50:31Z","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":"da413d1ada2601b78f8abf1b2e65d30ff929bc429ec0ea51c899e2da09159646","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-24T03:25:56Z","title_canon_sha256":"3d8c228d5434fbd556c0056d8bf7457bd7923f893b7fafa998a0be0e063466e7"},"schema_version":"1.0","source":{"id":"2502.16802","kind":"arxiv","version":3}},"canonical_sha256":"5669b20e8484f836551b64d6a9673264ab89584f6c4e5c5f1d6408f886de1e28","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5669b20e8484f836551b64d6a9673264ab89584f6c4e5c5f1d6408f886de1e28","first_computed_at":"2026-07-05T11:50:31.683577Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:50:31.683577Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"IdjvVkMFnafoOcKVyrNLKefFBy7v2nuU91D3f5DYNmIDX5BFjXY9dZr0bFIrLLtXXgmiBPcyqQArkgbDL97gAw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:50:31.684095Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.16802","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b6d019ee108fd11f3c194a9c021c3f4d77280d20ea71b51919d685f4b4042a37","sha256:71dca2c4575d0f495cf5604ae69f59d1d5040e72c378d147c7d84f109727b3e5"],"state_sha256":"f189e6ea8dd7db33cacebf3517a6516026d67930698cc0da6a4abdb89c359bb4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"k8rfVXpuh2ZfHkEqSNC0PBuPeiakMOLFwvmaPgasaFMAJ5gn9OVPMJ1ODHwdIgTioAb/ZHdh5PhmTjnYQ9h0Dw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-18T22:09:48.847319Z","bundle_sha256":"7592d8ef98e0f479a2d4c5dc94996204896a5e7467509ab943c70c129e30b470"}}