{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:SOSLLSMKZF4XK4TKKKW6G4ZS42","short_pith_number":"pith:SOSLLSMK","canonical_record":{"source":{"id":"2403.04652","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-07T16:52:49Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3b589a85e1def4b87232902bab81fb6c7fecf85a508022df8c11e5581cbc60d6","abstract_canon_sha256":"987eac431d1d8c77c04b3d8b43cdc72c99cc9e1702f809aa7ec6f77439a820f6"},"schema_version":"1.0"},"canonical_sha256":"93a4b5c98ac97975726a52ade37332e683c9e3f759e9e587f5c7a789ac9151c0","source":{"kind":"arxiv","id":"2403.04652","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.04652","created_at":"2026-07-05T10:03:24Z"},{"alias_kind":"arxiv_version","alias_value":"2403.04652v3","created_at":"2026-07-05T10:03:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.04652","created_at":"2026-07-05T10:03:24Z"},{"alias_kind":"pith_short_12","alias_value":"SOSLLSMKZF4X","created_at":"2026-07-05T10:03:24Z"},{"alias_kind":"pith_short_16","alias_value":"SOSLLSMKZF4XK4TK","created_at":"2026-07-05T10:03:24Z"},{"alias_kind":"pith_short_8","alias_value":"SOSLLSMK","created_at":"2026-07-05T10:03:24Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:SOSLLSMKZF4XK4TKKKW6G4ZS42","target":"record","payload":{"canonical_record":{"source":{"id":"2403.04652","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-07T16:52:49Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3b589a85e1def4b87232902bab81fb6c7fecf85a508022df8c11e5581cbc60d6","abstract_canon_sha256":"987eac431d1d8c77c04b3d8b43cdc72c99cc9e1702f809aa7ec6f77439a820f6"},"schema_version":"1.0"},"canonical_sha256":"93a4b5c98ac97975726a52ade37332e683c9e3f759e9e587f5c7a789ac9151c0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:03:24.338091Z","signature_b64":"vRiVKnIlR3cCwjXRIEqHf1tWn8+0c19/uXMs5p4gSXJx+0n7PmN0V83g9BAQ8nVL2NLQ3nEUj7ydyTtDjGNeDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"93a4b5c98ac97975726a52ade37332e683c9e3f759e9e587f5c7a789ac9151c0","last_reissued_at":"2026-07-05T10:03:24.337524Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:03:24.337524Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2403.04652","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":"2026-07-05T10:03:24Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rAxYsMEzMSgRyJMI5BsTcYHVtafrVJ9evci6KY28NdOUJvhgEEbTxOOhxQijCHNJnxWAkO/WSNJaA9hO0QDXBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-23T14:24:19.282760Z"},"content_sha256":"3a645e3c70666b98d59501a9d6e03f668a912e7b14f5358c1cd8293a9bcbccea","schema_version":"1.0","event_id":"sha256:3a645e3c70666b98d59501a9d6e03f668a912e7b14f5358c1cd8293a9bcbccea"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:SOSLLSMKZF4XK4TKKKW6G4ZS42","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Yi: Open Foundation Models by 01.AI","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"The Yi models achieve strong benchmark results primarily through high-quality data curation from a cascaded deduplication and filtering pipeline on 3.1 trillion tokens.","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"01.AI: Alex Young, Bei Chen, Chao Li, Chengen Huang, Ge Zhang, Guanwei Zhang, Guoyin Wang, Heng Li, Jiangcheng Zhu, Jianqun Chen, Jing Chang, Kaidong Yu, Pengcheng Nie, Peng Liu, Qiang Liu, Senbin Yang, Shawn Yue, Shiming Yang, Wenhao Huang, Wen Xie, Xiaohui Hu, Xiaoyi Ren, Xinyao Niu, Yanpeng Li, Yuchi Xu, Yudong Liu, Yue Wang, Yuxuan Cai, Zhenyu Gu, Zhiyuan Liu, Zonghong Dai","submitted_at":"2024-03-07T16:52:49Z","abstract_excerpt":"We introduce the Yi model family, a series of language and multimodal models that demonstrate strong multi-dimensional capabilities. The Yi model family is based on 6B and 34B pretrained language models, then we extend them to chat models, 200K long context models, depth-upscaled models, and vision-language models. Our base models achieve strong performance on a wide range of benchmarks like MMLU, and our finetuned chat models deliver strong human preference rate on major evaluation platforms like AlpacaEval and Chatbot Arena. Building upon our scalable super-computing infrastructure and the c"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We attribute the performance of Yi models primarily to its data quality resulting from our data-engineering efforts.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the cascaded deduplication and quality filtering pipeline produces data whose superiority is the primary driver of the reported benchmark gains, without detailed ablations or public dataset comparisons.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Yi models are 6B and 34B open foundation models pretrained on 3.1T curated tokens that achieve strong benchmark results through data quality and targeted extensions like long context and vision alignment.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"The Yi models achieve strong benchmark results primarily through high-quality data curation from a cascaded deduplication and filtering pipeline on 3.1 trillion tokens.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"52ac091abdc91890328fd6789d59e303135cd66872aae9a3b117bad20ea321e2"},"source":{"id":"2403.04652","kind":"arxiv","version":3},"verdict":{"id":"b0980a1f-216d-4190-9e2a-e3e0e6c1ae54","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-13T05:39:05.639941Z","strongest_claim":"We attribute the performance of Yi models primarily to its data quality resulting from our data-engineering efforts.","one_line_summary":"Yi models are 6B and 34B open foundation models pretrained on 3.1T curated tokens that achieve strong benchmark results through data quality and targeted extensions like long context and vision alignment.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the cascaded deduplication and quality filtering pipeline produces data whose superiority is the primary driver of the reported benchmark gains, without detailed ablations or public dataset comparisons.","pith_extraction_headline":"The Yi models achieve strong benchmark results primarily through high-quality data curation from a cascaded deduplication and filtering pipeline on 3.1 trillion tokens."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2403.04652/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":95,"sample":[{"doi":"","year":2023,"title":"GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints","work_id":"b73ad5b2-e553-4c71-b0c9-67e67ba7b158","ref_index":1,"cited_arxiv_id":"2305.13245","is_internal_anchor":true},{"doi":"","year":2021,"title":"Program Synthesis with Large Language Models","work_id":"fd241a05-03b9-4de2-9588-9d77ce176125","ref_index":2,"cited_arxiv_id":"2108.07732","is_internal_anchor":true},{"doi":"","year":null,"title":"Qwen Technical Report","work_id":"1c21f5f5-72d8-4d70-9def-40f6d89c208b","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Qwen Technical Report","work_id":"bb1fd52f-6b2f-437c-9516-37bdf6eb9be8","ref_index":4,"cited_arxiv_id":"2309.16609","is_internal_anchor":true},{"doi":"","year":1911,"title":"PIQA: Reasoning about Physical Commonsense in Natural Language","work_id":"0d865a62-6376-4606-8d3a-eeb3b6e9ba6d","ref_index":5,"cited_arxiv_id":"1911.11641","is_internal_anchor":true}],"resolved_work":95,"snapshot_sha256":"85a068666244f5c0d1db5928f5ec79925bdde36902d74e12b87227fd773f6afd","internal_anchors":32},"formal_canon":{"evidence_count":2,"snapshot_sha256":"00ab310f05fa5df9a21a2c66a74dd9e95fe63f11c40f9c25fe1da1631bd4d769"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"b0980a1f-216d-4190-9e2a-e3e0e6c1ae54"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:03:24Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"m8UcYQLymfGrWSBoH9UfDZzeBaqj7zFmO27rxrMmAXMYJrJIxDxtDKTfA3zxJqehoTYltenycWOD1VQp+NFRDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-23T14:24:19.283565Z"},"content_sha256":"3b6a19b811e2832b8ba34afd64d782f64401b7a06573631404be967b5c165064","schema_version":"1.0","event_id":"sha256:3b6a19b811e2832b8ba34afd64d782f64401b7a06573631404be967b5c165064"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SOSLLSMKZF4XK4TKKKW6G4ZS42/bundle.json","state_url":"https://pith.science/pith/SOSLLSMKZF4XK4TKKKW6G4ZS42/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SOSLLSMKZF4XK4TKKKW6G4ZS42/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":"2026-08-23T14:24:19Z","links":{"resolver":"https://pith.science/pith/SOSLLSMKZF4XK4TKKKW6G4ZS42","bundle":"https://pith.science/pith/SOSLLSMKZF4XK4TKKKW6G4ZS42/bundle.json","state":"https://pith.science/pith/SOSLLSMKZF4XK4TKKKW6G4ZS42/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SOSLLSMKZF4XK4TKKKW6G4ZS42/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:SOSLLSMKZF4XK4TKKKW6G4ZS42","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":"987eac431d1d8c77c04b3d8b43cdc72c99cc9e1702f809aa7ec6f77439a820f6","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-07T16:52:49Z","title_canon_sha256":"3b589a85e1def4b87232902bab81fb6c7fecf85a508022df8c11e5581cbc60d6"},"schema_version":"1.0","source":{"id":"2403.04652","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.04652","created_at":"2026-07-05T10:03:24Z"},{"alias_kind":"arxiv_version","alias_value":"2403.04652v3","created_at":"2026-07-05T10:03:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.04652","created_at":"2026-07-05T10:03:24Z"},{"alias_kind":"pith_short_12","alias_value":"SOSLLSMKZF4X","created_at":"2026-07-05T10:03:24Z"},{"alias_kind":"pith_short_16","alias_value":"SOSLLSMKZF4XK4TK","created_at":"2026-07-05T10:03:24Z"},{"alias_kind":"pith_short_8","alias_value":"SOSLLSMK","created_at":"2026-07-05T10:03:24Z"}],"graph_snapshots":[{"event_id":"sha256:3b6a19b811e2832b8ba34afd64d782f64401b7a06573631404be967b5c165064","target":"graph","created_at":"2026-07-05T10:03:24Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"We attribute the performance of Yi models primarily to its data quality resulting from our data-engineering efforts."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the cascaded deduplication and quality filtering pipeline produces data whose superiority is the primary driver of the reported benchmark gains, without detailed ablations or public dataset comparisons."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Yi models are 6B and 34B open foundation models pretrained on 3.1T curated tokens that achieve strong benchmark results through data quality and targeted extensions like long context and vision alignment."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"The Yi models achieve strong benchmark results primarily through high-quality data curation from a cascaded deduplication and filtering pipeline on 3.1 trillion tokens."}],"snapshot_sha256":"52ac091abdc91890328fd6789d59e303135cd66872aae9a3b117bad20ea321e2"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"00ab310f05fa5df9a21a2c66a74dd9e95fe63f11c40f9c25fe1da1631bd4d769"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2403.04652/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce the Yi model family, a series of language and multimodal models that demonstrate strong multi-dimensional capabilities. The Yi model family is based on 6B and 34B pretrained language models, then we extend them to chat models, 200K long context models, depth-upscaled models, and vision-language models. Our base models achieve strong performance on a wide range of benchmarks like MMLU, and our finetuned chat models deliver strong human preference rate on major evaluation platforms like AlpacaEval and Chatbot Arena. Building upon our scalable super-computing infrastructure and the c","authors_text":"01.AI: Alex Young, Bei Chen, Chao Li, Chengen Huang, Ge Zhang, Guanwei Zhang, Guoyin Wang, Heng Li, Jiangcheng Zhu, Jianqun Chen, Jing Chang, Kaidong Yu, Pengcheng Nie, Peng Liu, Qiang Liu, Senbin Yang, Shawn Yue, Shiming Yang, Wenhao Huang, Wen Xie, Xiaohui Hu, Xiaoyi Ren, Xinyao Niu, Yanpeng Li, Yuchi Xu, Yudong Liu, Yue Wang, Yuxuan Cai, Zhenyu Gu, Zhiyuan Liu, Zonghong Dai","cross_cats":["cs.AI"],"headline":"The Yi models achieve strong benchmark results primarily through high-quality data curation from a cascaded deduplication and filtering pipeline on 3.1 trillion tokens.","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-07T16:52:49Z","title":"Yi: Open Foundation Models by 01.AI"},"references":{"count":95,"internal_anchors":32,"resolved_work":95,"sample":[{"cited_arxiv_id":"2305.13245","doi":"","is_internal_anchor":true,"ref_index":1,"title":"GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints","work_id":"b73ad5b2-e553-4c71-b0c9-67e67ba7b158","year":2023},{"cited_arxiv_id":"2108.07732","doi":"","is_internal_anchor":true,"ref_index":2,"title":"Program Synthesis with Large Language Models","work_id":"fd241a05-03b9-4de2-9588-9d77ce176125","year":2021},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Qwen Technical Report","work_id":"1c21f5f5-72d8-4d70-9def-40f6d89c208b","year":null},{"cited_arxiv_id":"2309.16609","doi":"","is_internal_anchor":true,"ref_index":4,"title":"Qwen Technical Report","work_id":"bb1fd52f-6b2f-437c-9516-37bdf6eb9be8","year":null},{"cited_arxiv_id":"1911.11641","doi":"","is_internal_anchor":true,"ref_index":5,"title":"PIQA: Reasoning about Physical Commonsense in Natural Language","work_id":"0d865a62-6376-4606-8d3a-eeb3b6e9ba6d","year":1911}],"snapshot_sha256":"85a068666244f5c0d1db5928f5ec79925bdde36902d74e12b87227fd773f6afd"},"source":{"id":"2403.04652","kind":"arxiv","version":3},"verdict":{"created_at":"2026-05-13T05:39:05.639941Z","id":"b0980a1f-216d-4190-9e2a-e3e0e6c1ae54","model_set":{"reader":"grok-4.3"},"one_line_summary":"Yi models are 6B and 34B open foundation models pretrained on 3.1T curated tokens that achieve strong benchmark results through data quality and targeted extensions like long context and vision alignment.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"The Yi models achieve strong benchmark results primarily through high-quality data curation from a cascaded deduplication and filtering pipeline on 3.1 trillion tokens.","strongest_claim":"We attribute the performance of Yi models primarily to its data quality resulting from our data-engineering efforts.","weakest_assumption":"That the cascaded deduplication and quality filtering pipeline produces data whose superiority is the primary driver of the reported benchmark gains, without detailed ablations or public dataset comparisons."}},"verdict_id":"b0980a1f-216d-4190-9e2a-e3e0e6c1ae54"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:3a645e3c70666b98d59501a9d6e03f668a912e7b14f5358c1cd8293a9bcbccea","target":"record","created_at":"2026-07-05T10:03:24Z","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":"987eac431d1d8c77c04b3d8b43cdc72c99cc9e1702f809aa7ec6f77439a820f6","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-07T16:52:49Z","title_canon_sha256":"3b589a85e1def4b87232902bab81fb6c7fecf85a508022df8c11e5581cbc60d6"},"schema_version":"1.0","source":{"id":"2403.04652","kind":"arxiv","version":3}},"canonical_sha256":"93a4b5c98ac97975726a52ade37332e683c9e3f759e9e587f5c7a789ac9151c0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"93a4b5c98ac97975726a52ade37332e683c9e3f759e9e587f5c7a789ac9151c0","first_computed_at":"2026-07-05T10:03:24.337524Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:03:24.337524Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"vRiVKnIlR3cCwjXRIEqHf1tWn8+0c19/uXMs5p4gSXJx+0n7PmN0V83g9BAQ8nVL2NLQ3nEUj7ydyTtDjGNeDw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:03:24.338091Z","signed_message":"canonical_sha256_bytes"},"source_id":"2403.04652","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3a645e3c70666b98d59501a9d6e03f668a912e7b14f5358c1cd8293a9bcbccea","sha256:3b6a19b811e2832b8ba34afd64d782f64401b7a06573631404be967b5c165064"],"state_sha256":"8521d9e7d9a28aad18df250553f48701c3c1a7732451533c95537358e610172c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"60V24zqGENxPhvF8UwLuDQ3VHSuNphZOnayDeKNEdqHvO02e5fDZYV65uRWWd7tNjQLr/zmnq215QUiWDBgqBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-23T14:24:19.289589Z","bundle_sha256":"1691cf82187ce3d031e48723dc468997491c68a0e4a3a9a3f79b2bdda501e2b9"}}