{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:2K3ABXNKJR5K4LO5FI56LLTKEP","short_pith_number":"pith:2K3ABXNK","canonical_record":{"source":{"id":"2206.10789","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-06-22T01:11:29Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"571c23f23fbfa8348c5ecbead6d7f0fd765d74d597c5dafef77159b3c18eb1fe","abstract_canon_sha256":"0f8a03a77457910165202ec1f2c5372f66bb10786cb4fdf974dc2829b4423959"},"schema_version":"1.0"},"canonical_sha256":"d2b600ddaa4c7aae2ddd2a3be5ae6a23d7de8d9858cc3e7df8b75ab81c903c09","source":{"kind":"arxiv","id":"2206.10789","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2206.10789","created_at":"2026-07-05T04:33:55Z"},{"alias_kind":"arxiv_version","alias_value":"2206.10789v1","created_at":"2026-07-05T04:33:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.10789","created_at":"2026-07-05T04:33:55Z"},{"alias_kind":"pith_short_12","alias_value":"2K3ABXNKJR5K","created_at":"2026-07-05T04:33:55Z"},{"alias_kind":"pith_short_16","alias_value":"2K3ABXNKJR5K4LO5","created_at":"2026-07-05T04:33:55Z"},{"alias_kind":"pith_short_8","alias_value":"2K3ABXNK","created_at":"2026-07-05T04:33:55Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:2K3ABXNKJR5K4LO5FI56LLTKEP","target":"record","payload":{"canonical_record":{"source":{"id":"2206.10789","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-06-22T01:11:29Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"571c23f23fbfa8348c5ecbead6d7f0fd765d74d597c5dafef77159b3c18eb1fe","abstract_canon_sha256":"0f8a03a77457910165202ec1f2c5372f66bb10786cb4fdf974dc2829b4423959"},"schema_version":"1.0"},"canonical_sha256":"d2b600ddaa4c7aae2ddd2a3be5ae6a23d7de8d9858cc3e7df8b75ab81c903c09","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:33:55.558963Z","signature_b64":"B3yfCuwyTGKiijMciUS1FVPMKwffAzsaoyRb+UgxYUCku5hIhA/i+w/gqVWc9KyVZ4M82YWqZ0Ct1JYC19CiDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d2b600ddaa4c7aae2ddd2a3be5ae6a23d7de8d9858cc3e7df8b75ab81c903c09","last_reissued_at":"2026-07-05T04:33:55.558434Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:33:55.558434Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2206.10789","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":"2026-07-05T04:33:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"v9sKjXyKitbrPBW+nLoc2FqmX3DzaLoy6sIjSlLinqpyFSCBqvCAO/GotOzBCU4EpQU61+hH3InIfZ73u8WwAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T01:06:49.826515Z"},"content_sha256":"1aed029c3943b24ba0a356af2d1134e20b1fe7a1b962a5a690c92a805c743101","schema_version":"1.0","event_id":"sha256:1aed029c3943b24ba0a356af2d1134e20b1fe7a1b962a5a690c92a805c743101"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:2K3ABXNKJR5K4LO5FI56LLTKEP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Scaling Autoregressive Models for Content-Rich Text-to-Image Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Scaling an autoregressive encoder-decoder Transformer to 20 billion parameters produces high-fidelity images from text prompts that include complex compositions and world knowledge.","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Alexander Ku, Ben Hutchinson, Burcu Karagol Ayan, Gunjan Baid, Han Zhang, Jason Baldridge, Jiahui Yu, Jing Yu Koh, Thang Luong, Vijay Vasudevan, Wei Han, Xin Li, Yinfei Yang, Yonghui Wu, Yuanzhong Xu, Zarana Parekh, Zirui Wang","submitted_at":"2022-06-22T01:11:29Z","abstract_excerpt":"We present the Pathways Autoregressive Text-to-Image (Parti) model, which generates high-fidelity photorealistic images and supports content-rich synthesis involving complex compositions and world knowledge. Parti treats text-to-image generation as a sequence-to-sequence modeling problem, akin to machine translation, with sequences of image tokens as the target outputs rather than text tokens in another language. This strategy can naturally tap into the rich body of prior work on large language models, which have seen continued advances in capabilities and performance through scaling data and "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"we achieve consistent quality improvements by scaling the encoder-decoder Transformer model up to 20B parameters, with a new state-of-the-art zero-shot FID score of 7.23 and finetuned FID score of 3.22 on MS-COCO.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the ViT-VQGAN tokenizer preserves enough visual information for high-fidelity reconstruction and that further scaling will continue to improve quality without new failure modes on complex compositions.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Scaling an autoregressive Transformer to 20B parameters for text-to-image generation using image token sequences achieves new SOTA zero-shot FID of 7.23 and fine-tuned FID of 3.22 on MS-COCO.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Scaling an autoregressive encoder-decoder Transformer to 20 billion parameters produces high-fidelity images from text prompts that include complex compositions and world knowledge.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"178e53282adee1d37d3bf7baa2494ab5982ccf0cd818dfe8b6320a936f230165"},"source":{"id":"2206.10789","kind":"arxiv","version":1},"verdict":{"id":"fe88d836-b3ef-44b7-ad45-baacb50cf016","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-12T04:45:30.299538Z","strongest_claim":"we achieve consistent quality improvements by scaling the encoder-decoder Transformer model up to 20B parameters, with a new state-of-the-art zero-shot FID score of 7.23 and finetuned FID score of 3.22 on MS-COCO.","one_line_summary":"Scaling an autoregressive Transformer to 20B parameters for text-to-image generation using image token sequences achieves new SOTA zero-shot FID of 7.23 and fine-tuned FID of 3.22 on MS-COCO.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the ViT-VQGAN tokenizer preserves enough visual information for high-fidelity reconstruction and that further scaling will continue to improve quality without new failure modes on complex compositions.","pith_extraction_headline":"Scaling an autoregressive encoder-decoder Transformer to 20 billion parameters produces high-fidelity images from text prompts that include complex compositions and world knowledge."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2206.10789/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":99,"sample":[{"doi":"","year":2021,"title":"Introducing pathways: A next-generation ai architecture","work_id":"6a44fc3b-2ced-4e68-bc47-c63f283c84b4","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"Zero-shot text-to-image generation","work_id":"cafdc74f-7fef-4b7a-a8a5-90ff5e1734b8","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"Cogview: Mastering text-to-image generation via transformers","work_id":"a2fa38fb-6d9f-4029-ba37-e2b157530dde","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2017,"title":"Attention is all you need","work_id":"ac9d72e5-eb60-417e-963d-25671207074e","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2016,"title":"Discrete Variational Autoencoders","work_id":"6dd92474-cf3a-4e7e-b5a6-6b52d48d57f2","ref_index":5,"cited_arxiv_id":"1609.02200","is_internal_anchor":false}],"resolved_work":99,"snapshot_sha256":"554bddb1642904b0ac700818711e4e9d70c507e24a8bb84ac43a488a50511e1a","internal_anchors":18},"formal_canon":{"evidence_count":2,"snapshot_sha256":"114ca4e6a534272b5caad9c91bd2df38e6a805b843863629e21d6bf64950c25a"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"fe88d836-b3ef-44b7-ad45-baacb50cf016"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T04:33:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"w1zst3GDQmHPdmrqq22xBbPFBYCgh0Irzx0LQAh2sewyxK5D1tiDJRjhWLYv5TpxikfILbGtR3lRR3OVZ8wWDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T01:06:49.827378Z"},"content_sha256":"2d1792424b6da01fdfea99b06f89af2fc4c6a08857e23d1d60e6b9770aaa8f8b","schema_version":"1.0","event_id":"sha256:2d1792424b6da01fdfea99b06f89af2fc4c6a08857e23d1d60e6b9770aaa8f8b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/2K3ABXNKJR5K4LO5FI56LLTKEP/bundle.json","state_url":"https://pith.science/pith/2K3ABXNKJR5K4LO5FI56LLTKEP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/2K3ABXNKJR5K4LO5FI56LLTKEP/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":"2026-08-11T01:06:49Z","links":{"resolver":"https://pith.science/pith/2K3ABXNKJR5K4LO5FI56LLTKEP","bundle":"https://pith.science/pith/2K3ABXNKJR5K4LO5FI56LLTKEP/bundle.json","state":"https://pith.science/pith/2K3ABXNKJR5K4LO5FI56LLTKEP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/2K3ABXNKJR5K4LO5FI56LLTKEP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:2K3ABXNKJR5K4LO5FI56LLTKEP","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":"0f8a03a77457910165202ec1f2c5372f66bb10786cb4fdf974dc2829b4423959","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-06-22T01:11:29Z","title_canon_sha256":"571c23f23fbfa8348c5ecbead6d7f0fd765d74d597c5dafef77159b3c18eb1fe"},"schema_version":"1.0","source":{"id":"2206.10789","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2206.10789","created_at":"2026-07-05T04:33:55Z"},{"alias_kind":"arxiv_version","alias_value":"2206.10789v1","created_at":"2026-07-05T04:33:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.10789","created_at":"2026-07-05T04:33:55Z"},{"alias_kind":"pith_short_12","alias_value":"2K3ABXNKJR5K","created_at":"2026-07-05T04:33:55Z"},{"alias_kind":"pith_short_16","alias_value":"2K3ABXNKJR5K4LO5","created_at":"2026-07-05T04:33:55Z"},{"alias_kind":"pith_short_8","alias_value":"2K3ABXNK","created_at":"2026-07-05T04:33:55Z"}],"graph_snapshots":[{"event_id":"sha256:2d1792424b6da01fdfea99b06f89af2fc4c6a08857e23d1d60e6b9770aaa8f8b","target":"graph","created_at":"2026-07-05T04:33:55Z","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 achieve consistent quality improvements by scaling the encoder-decoder Transformer model up to 20B parameters, with a new state-of-the-art zero-shot FID score of 7.23 and finetuned FID score of 3.22 on MS-COCO."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the ViT-VQGAN tokenizer preserves enough visual information for high-fidelity reconstruction and that further scaling will continue to improve quality without new failure modes on complex compositions."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Scaling an autoregressive Transformer to 20B parameters for text-to-image generation using image token sequences achieves new SOTA zero-shot FID of 7.23 and fine-tuned FID of 3.22 on MS-COCO."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Scaling an autoregressive encoder-decoder Transformer to 20 billion parameters produces high-fidelity images from text prompts that include complex compositions and world knowledge."}],"snapshot_sha256":"178e53282adee1d37d3bf7baa2494ab5982ccf0cd818dfe8b6320a936f230165"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"114ca4e6a534272b5caad9c91bd2df38e6a805b843863629e21d6bf64950c25a"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2206.10789/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We present the Pathways Autoregressive Text-to-Image (Parti) model, which generates high-fidelity photorealistic images and supports content-rich synthesis involving complex compositions and world knowledge. Parti treats text-to-image generation as a sequence-to-sequence modeling problem, akin to machine translation, with sequences of image tokens as the target outputs rather than text tokens in another language. This strategy can naturally tap into the rich body of prior work on large language models, which have seen continued advances in capabilities and performance through scaling data and ","authors_text":"Alexander Ku, Ben Hutchinson, Burcu Karagol Ayan, Gunjan Baid, Han Zhang, Jason Baldridge, Jiahui Yu, Jing Yu Koh, Thang Luong, Vijay Vasudevan, Wei Han, Xin Li, Yinfei Yang, Yonghui Wu, Yuanzhong Xu, Zarana Parekh, Zirui Wang","cross_cats":["cs.LG"],"headline":"Scaling an autoregressive encoder-decoder Transformer to 20 billion parameters produces high-fidelity images from text prompts that include complex compositions and world knowledge.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-06-22T01:11:29Z","title":"Scaling Autoregressive Models for Content-Rich Text-to-Image Generation"},"references":{"count":99,"internal_anchors":18,"resolved_work":99,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Introducing pathways: A next-generation ai architecture","work_id":"6a44fc3b-2ced-4e68-bc47-c63f283c84b4","year":2021},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Zero-shot text-to-image generation","work_id":"cafdc74f-7fef-4b7a-a8a5-90ff5e1734b8","year":2021},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Cogview: Mastering text-to-image generation via transformers","work_id":"a2fa38fb-6d9f-4029-ba37-e2b157530dde","year":2021},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Attention is all you need","work_id":"ac9d72e5-eb60-417e-963d-25671207074e","year":2017},{"cited_arxiv_id":"1609.02200","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Discrete Variational Autoencoders","work_id":"6dd92474-cf3a-4e7e-b5a6-6b52d48d57f2","year":2016}],"snapshot_sha256":"554bddb1642904b0ac700818711e4e9d70c507e24a8bb84ac43a488a50511e1a"},"source":{"id":"2206.10789","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-12T04:45:30.299538Z","id":"fe88d836-b3ef-44b7-ad45-baacb50cf016","model_set":{"reader":"grok-4.3"},"one_line_summary":"Scaling an autoregressive Transformer to 20B parameters for text-to-image generation using image token sequences achieves new SOTA zero-shot FID of 7.23 and fine-tuned FID of 3.22 on MS-COCO.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Scaling an autoregressive encoder-decoder Transformer to 20 billion parameters produces high-fidelity images from text prompts that include complex compositions and world knowledge.","strongest_claim":"we achieve consistent quality improvements by scaling the encoder-decoder Transformer model up to 20B parameters, with a new state-of-the-art zero-shot FID score of 7.23 and finetuned FID score of 3.22 on MS-COCO.","weakest_assumption":"That the ViT-VQGAN tokenizer preserves enough visual information for high-fidelity reconstruction and that further scaling will continue to improve quality without new failure modes on complex compositions."}},"verdict_id":"fe88d836-b3ef-44b7-ad45-baacb50cf016"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:1aed029c3943b24ba0a356af2d1134e20b1fe7a1b962a5a690c92a805c743101","target":"record","created_at":"2026-07-05T04:33:55Z","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":"0f8a03a77457910165202ec1f2c5372f66bb10786cb4fdf974dc2829b4423959","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-06-22T01:11:29Z","title_canon_sha256":"571c23f23fbfa8348c5ecbead6d7f0fd765d74d597c5dafef77159b3c18eb1fe"},"schema_version":"1.0","source":{"id":"2206.10789","kind":"arxiv","version":1}},"canonical_sha256":"d2b600ddaa4c7aae2ddd2a3be5ae6a23d7de8d9858cc3e7df8b75ab81c903c09","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d2b600ddaa4c7aae2ddd2a3be5ae6a23d7de8d9858cc3e7df8b75ab81c903c09","first_computed_at":"2026-07-05T04:33:55.558434Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T04:33:55.558434Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"B3yfCuwyTGKiijMciUS1FVPMKwffAzsaoyRb+UgxYUCku5hIhA/i+w/gqVWc9KyVZ4M82YWqZ0Ct1JYC19CiDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T04:33:55.558963Z","signed_message":"canonical_sha256_bytes"},"source_id":"2206.10789","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1aed029c3943b24ba0a356af2d1134e20b1fe7a1b962a5a690c92a805c743101","sha256:2d1792424b6da01fdfea99b06f89af2fc4c6a08857e23d1d60e6b9770aaa8f8b"],"state_sha256":"0288f3541a5f106bb46cdcff5d41707967e6663063ff54f9a4e8d7d704d471ff"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sM9a9Acsyoj94DWsczeaarZLSEJuvvOPGzgMNJVwsyyuKkx5hQ7JpNKShFMdSpm0vA+UaMkJHPVU1nziVeYcCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-11T01:06:49.832109Z","bundle_sha256":"74fa2337dfb9ed82fad82338e1b5435d829726e98aa2d10526076b0d0e87c277"}}