{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:WT2B2WJP3OXQYVOSO2KO6W37AP","short_pith_number":"pith:WT2B2WJP","canonical_record":{"source":{"id":"2512.24120","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-12-30T10:01:55Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"cef0f864368c9e74340f02a0e0288eeb6143455ece25681089161a2f1d5d8acc","abstract_canon_sha256":"035e2a7307fd7baaf807caed199e57d0ff222816b1e5b2d57d969f5c02244eee"},"schema_version":"1.0"},"canonical_sha256":"b4f41d592fdbaf0c55d27694ef5b7f03f03e6f91c60d07bb18897263dc298645","source":{"kind":"arxiv","id":"2512.24120","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2512.24120","created_at":"2026-06-02T02:04:12Z"},{"alias_kind":"arxiv_version","alias_value":"2512.24120v2","created_at":"2026-06-02T02:04:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.24120","created_at":"2026-06-02T02:04:12Z"},{"alias_kind":"pith_short_12","alias_value":"WT2B2WJP3OXQ","created_at":"2026-06-02T02:04:12Z"},{"alias_kind":"pith_short_16","alias_value":"WT2B2WJP3OXQYVOS","created_at":"2026-06-02T02:04:12Z"},{"alias_kind":"pith_short_8","alias_value":"WT2B2WJP","created_at":"2026-06-02T02:04:12Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:WT2B2WJP3OXQYVOSO2KO6W37AP","target":"record","payload":{"canonical_record":{"source":{"id":"2512.24120","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-12-30T10:01:55Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"cef0f864368c9e74340f02a0e0288eeb6143455ece25681089161a2f1d5d8acc","abstract_canon_sha256":"035e2a7307fd7baaf807caed199e57d0ff222816b1e5b2d57d969f5c02244eee"},"schema_version":"1.0"},"canonical_sha256":"b4f41d592fdbaf0c55d27694ef5b7f03f03e6f91c60d07bb18897263dc298645","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T02:04:12.786625Z","signature_b64":"chSAGmKytDhR/jGqW45LoKkb56IzY8jmWI+nIJEty8UIpXtH4EPTr/NvEnoxlTMsApnIjnkz/gB7/Su+Ap/XBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b4f41d592fdbaf0c55d27694ef5b7f03f03e6f91c60d07bb18897263dc298645","last_reissued_at":"2026-06-02T02:04:12.786101Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T02:04:12.786101Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2512.24120","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":"2026-06-02T02:04:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"EHF089OEwPnU23R0u160vDEqcJmxLEnvzSzMLg8Zqfl22byGPOvJzj2ezV/birQXqkuRAsLrnJ5DJLywTbsFCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-08T14:06:46.827062Z"},"content_sha256":"891f96cb7caf9fb7381af0fd37d17dc8f4cc296f104e37af4cde2b1319a1b395","schema_version":"1.0","event_id":"sha256:891f96cb7caf9fb7381af0fd37d17dc8f4cc296f104e37af4cde2b1319a1b395"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:WT2B2WJP3OXQYVOSO2KO6W37AP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Enhancing LLM-Based Neural Network Generation: Few-Shot Prompting and Efficient Validation for Automated Architecture Design","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Three examples in few-shot prompts let LLMs generate the most balanced neural architectures for vision tasks while a simple hash check speeds validation by 100 times.","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Avi Goyal, Chandini Vysyaraju, Dmitry Ignatov, Radu Timofte, Raghuvir Duvvuri","submitted_at":"2025-12-30T10:01:55Z","abstract_excerpt":"Automated neural network architecture design remains a significant challenge in computer vision. Task diversity and computational constraints require both effective architectures and efficient search methods. Large Language Models (LLMs) present a promising alternative to computationally intensive Neural Architecture Search (NAS), but their application to architecture generation in computer vision has not been systematically studied, particularly regarding prompt engineering and validation strategies. Building on the task-agnostic NNGPT/LEMUR framework, this work introduces and validates two k"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Using n = 3 examples best balances architectural diversity and context focus for vision tasks. Whitespace-Normalized Hash Validation provides a 100x speedup over AST parsing and prevents redundant training of duplicate architectures.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the observed performance differences across prompting regimes are driven by the number of examples rather than by uncontrolled variation in LLM sampling, dataset splits, or training hyperparameters.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Three-example few-shot prompting optimizes LLM-generated vision architectures while a whitespace-normalized hash provides 100x faster duplicate detection than AST parsing across seven benchmarks.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Three examples in few-shot prompts let LLMs generate the most balanced neural architectures for vision tasks while a simple hash check speeds validation by 100 times.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"cd957f11bddfb375c8dc6beceae12548ee9b1bf94315874b5e62529def1fe952"},"source":{"id":"2512.24120","kind":"arxiv","version":2},"verdict":{"id":"b37b81dd-8def-446a-bef4-eadc73a2c45e","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T19:04:48.401714Z","strongest_claim":"Using n = 3 examples best balances architectural diversity and context focus for vision tasks. Whitespace-Normalized Hash Validation provides a 100x speedup over AST parsing and prevents redundant training of duplicate architectures.","one_line_summary":"Three-example few-shot prompting optimizes LLM-generated vision architectures while a whitespace-normalized hash provides 100x faster duplicate detection than AST parsing across seven benchmarks.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the observed performance differences across prompting regimes are driven by the number of examples rather than by uncontrolled variation in LLM sampling, dataset splits, or training hyperparameters.","pith_extraction_headline":"Three examples in few-shot prompts let LLMs generate the most balanced neural architectures for vision tasks while a simple hash check speeds validation by 100 times."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2512.24120/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":30,"sample":[{"doi":"","year":2025,"title":"AUGMENTGEST: Can random data cropping augmentation boost gesture recognition performance?arXiv preprint","work_id":"9e0f2dfb-060e-4bd6-864f-570f7d6e05bd","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1901,"title":"Brown, Benjamin Mann, Nick Ryder, Melanie Sub- biah, Jared Kaplan, Prafulla Dhariwal, et al","work_id":"b79516a7-0610-419f-9be6-aea40d989bd0","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"Evaluating Large Language Models Trained on Code","work_id":"042493e9-b26f-4b4e-bbde-382072ca9b08","ref_index":3,"cited_arxiv_id":"2107.03374","is_internal_anchor":true},{"doi":"","year":2022,"title":"Program Synthesis with Large Language Models","work_id":"fd241a05-03b9-4de2-9588-9d77ce176125","ref_index":4,"cited_arxiv_id":"2108.07732","is_internal_anchor":true},{"doi":"","year":2024,"title":"DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligence","work_id":"f22dae5a-27e2-41d0-a061-c4286418dee3","ref_index":5,"cited_arxiv_id":"2401.14196","is_internal_anchor":true}],"resolved_work":30,"snapshot_sha256":"17a585f11edc9b258b28669d64abaf33ba4382e534ddf4664e92b4934377ec72","internal_anchors":8},"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":"b37b81dd-8def-446a-bef4-eadc73a2c45e"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-02T02:04:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"oQwwTWz7amdq2DlavCJJe3iIyIIDD8XSIc5XaNrL3bf3Rl5X9L9LnWxsNC5CaT4sL/WlwL5XczwwgmgdlKeRDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-08T14:06:46.827935Z"},"content_sha256":"f3a24e65b58f5a14578aca887034f435c9b53b3332fbf5a3437346c94357912e","schema_version":"1.0","event_id":"sha256:f3a24e65b58f5a14578aca887034f435c9b53b3332fbf5a3437346c94357912e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/WT2B2WJP3OXQYVOSO2KO6W37AP/bundle.json","state_url":"https://pith.science/pith/WT2B2WJP3OXQYVOSO2KO6W37AP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/WT2B2WJP3OXQYVOSO2KO6W37AP/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":"2026-06-08T14:06:46Z","links":{"resolver":"https://pith.science/pith/WT2B2WJP3OXQYVOSO2KO6W37AP","bundle":"https://pith.science/pith/WT2B2WJP3OXQYVOSO2KO6W37AP/bundle.json","state":"https://pith.science/pith/WT2B2WJP3OXQYVOSO2KO6W37AP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/WT2B2WJP3OXQYVOSO2KO6W37AP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:WT2B2WJP3OXQYVOSO2KO6W37AP","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":"035e2a7307fd7baaf807caed199e57d0ff222816b1e5b2d57d969f5c02244eee","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-12-30T10:01:55Z","title_canon_sha256":"cef0f864368c9e74340f02a0e0288eeb6143455ece25681089161a2f1d5d8acc"},"schema_version":"1.0","source":{"id":"2512.24120","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2512.24120","created_at":"2026-06-02T02:04:12Z"},{"alias_kind":"arxiv_version","alias_value":"2512.24120v2","created_at":"2026-06-02T02:04:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.24120","created_at":"2026-06-02T02:04:12Z"},{"alias_kind":"pith_short_12","alias_value":"WT2B2WJP3OXQ","created_at":"2026-06-02T02:04:12Z"},{"alias_kind":"pith_short_16","alias_value":"WT2B2WJP3OXQYVOS","created_at":"2026-06-02T02:04:12Z"},{"alias_kind":"pith_short_8","alias_value":"WT2B2WJP","created_at":"2026-06-02T02:04:12Z"}],"graph_snapshots":[{"event_id":"sha256:f3a24e65b58f5a14578aca887034f435c9b53b3332fbf5a3437346c94357912e","target":"graph","created_at":"2026-06-02T02:04:12Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Using n = 3 examples best balances architectural diversity and context focus for vision tasks. Whitespace-Normalized Hash Validation provides a 100x speedup over AST parsing and prevents redundant training of duplicate architectures."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the observed performance differences across prompting regimes are driven by the number of examples rather than by uncontrolled variation in LLM sampling, dataset splits, or training hyperparameters."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Three-example few-shot prompting optimizes LLM-generated vision architectures while a whitespace-normalized hash provides 100x faster duplicate detection than AST parsing across seven benchmarks."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Three examples in few-shot prompts let LLMs generate the most balanced neural architectures for vision tasks while a simple hash check speeds validation by 100 times."}],"snapshot_sha256":"cd957f11bddfb375c8dc6beceae12548ee9b1bf94315874b5e62529def1fe952"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2512.24120/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Automated neural network architecture design remains a significant challenge in computer vision. Task diversity and computational constraints require both effective architectures and efficient search methods. Large Language Models (LLMs) present a promising alternative to computationally intensive Neural Architecture Search (NAS), but their application to architecture generation in computer vision has not been systematically studied, particularly regarding prompt engineering and validation strategies. Building on the task-agnostic NNGPT/LEMUR framework, this work introduces and validates two k","authors_text":"Avi Goyal, Chandini Vysyaraju, Dmitry Ignatov, Radu Timofte, Raghuvir Duvvuri","cross_cats":["cs.AI"],"headline":"Three examples in few-shot prompts let LLMs generate the most balanced neural architectures for vision tasks while a simple hash check speeds validation by 100 times.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-12-30T10:01:55Z","title":"Enhancing LLM-Based Neural Network Generation: Few-Shot Prompting and Efficient Validation for Automated Architecture Design"},"references":{"count":30,"internal_anchors":8,"resolved_work":30,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"AUGMENTGEST: Can random data cropping augmentation boost gesture recognition performance?arXiv preprint","work_id":"9e0f2dfb-060e-4bd6-864f-570f7d6e05bd","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Brown, Benjamin Mann, Nick Ryder, Melanie Sub- biah, Jared Kaplan, Prafulla Dhariwal, et al","work_id":"b79516a7-0610-419f-9be6-aea40d989bd0","year":1901},{"cited_arxiv_id":"2107.03374","doi":"","is_internal_anchor":true,"ref_index":3,"title":"Evaluating Large Language Models Trained on Code","work_id":"042493e9-b26f-4b4e-bbde-382072ca9b08","year":2021},{"cited_arxiv_id":"2108.07732","doi":"","is_internal_anchor":true,"ref_index":4,"title":"Program Synthesis with Large Language Models","work_id":"fd241a05-03b9-4de2-9588-9d77ce176125","year":2022},{"cited_arxiv_id":"2401.14196","doi":"","is_internal_anchor":true,"ref_index":5,"title":"DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligence","work_id":"f22dae5a-27e2-41d0-a061-c4286418dee3","year":2024}],"snapshot_sha256":"17a585f11edc9b258b28669d64abaf33ba4382e534ddf4664e92b4934377ec72"},"source":{"id":"2512.24120","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-16T19:04:48.401714Z","id":"b37b81dd-8def-446a-bef4-eadc73a2c45e","model_set":{"reader":"grok-4.3"},"one_line_summary":"Three-example few-shot prompting optimizes LLM-generated vision architectures while a whitespace-normalized hash provides 100x faster duplicate detection than AST parsing across seven benchmarks.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Three examples in few-shot prompts let LLMs generate the most balanced neural architectures for vision tasks while a simple hash check speeds validation by 100 times.","strongest_claim":"Using n = 3 examples best balances architectural diversity and context focus for vision tasks. Whitespace-Normalized Hash Validation provides a 100x speedup over AST parsing and prevents redundant training of duplicate architectures.","weakest_assumption":"That the observed performance differences across prompting regimes are driven by the number of examples rather than by uncontrolled variation in LLM sampling, dataset splits, or training hyperparameters."}},"verdict_id":"b37b81dd-8def-446a-bef4-eadc73a2c45e"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:891f96cb7caf9fb7381af0fd37d17dc8f4cc296f104e37af4cde2b1319a1b395","target":"record","created_at":"2026-06-02T02:04:12Z","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":"035e2a7307fd7baaf807caed199e57d0ff222816b1e5b2d57d969f5c02244eee","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-12-30T10:01:55Z","title_canon_sha256":"cef0f864368c9e74340f02a0e0288eeb6143455ece25681089161a2f1d5d8acc"},"schema_version":"1.0","source":{"id":"2512.24120","kind":"arxiv","version":2}},"canonical_sha256":"b4f41d592fdbaf0c55d27694ef5b7f03f03e6f91c60d07bb18897263dc298645","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b4f41d592fdbaf0c55d27694ef5b7f03f03e6f91c60d07bb18897263dc298645","first_computed_at":"2026-06-02T02:04:12.786101Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T02:04:12.786101Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"chSAGmKytDhR/jGqW45LoKkb56IzY8jmWI+nIJEty8UIpXtH4EPTr/NvEnoxlTMsApnIjnkz/gB7/Su+Ap/XBA==","signature_status":"signed_v1","signed_at":"2026-06-02T02:04:12.786625Z","signed_message":"canonical_sha256_bytes"},"source_id":"2512.24120","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:891f96cb7caf9fb7381af0fd37d17dc8f4cc296f104e37af4cde2b1319a1b395","sha256:f3a24e65b58f5a14578aca887034f435c9b53b3332fbf5a3437346c94357912e"],"state_sha256":"4d5869761c4e52a168e14e47e5fa974cfd9df12eba1dcf263e27c26a2d2a658a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"a683CViY/Yf0B6Xvwf2LG80Y/UtHTH3LccFeAN46Bhz0GSRso2daEHp+smH+RTzRu0FfP4VCQenJKYaXxb8QCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-08T14:06:46.831613Z","bundle_sha256":"e74150d088907b8139bc384802f46ee37d7c488ac4955de1fe9fdabf6f2e630c"}}