{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:WTB52ZRBMY3ZABR43EZHIM2ZMO","short_pith_number":"pith:WTB52ZRB","canonical_record":{"source":{"id":"2605.14117","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-13T21:06:12Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d81df676a297d8a4be1bb03038b110a20ce3989b64b91dc74cb62766389197ee","abstract_canon_sha256":"41f82cd9c756bc9a80d2f669439e4d1a14438dcbdc180145e816a33b2bc2636b"},"schema_version":"1.0"},"canonical_sha256":"b4c3dd6621663790063cd9327433596398c255e45a3ecdf70d19a7ba60f23337","source":{"kind":"arxiv","id":"2605.14117","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14117","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14117v1","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14117","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"pith_short_12","alias_value":"WTB52ZRBMY3Z","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"WTB52ZRBMY3ZABR4","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"WTB52ZRB","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:WTB52ZRBMY3ZABR43EZHIM2ZMO","target":"record","payload":{"canonical_record":{"source":{"id":"2605.14117","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-13T21:06:12Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d81df676a297d8a4be1bb03038b110a20ce3989b64b91dc74cb62766389197ee","abstract_canon_sha256":"41f82cd9c756bc9a80d2f669439e4d1a14438dcbdc180145e816a33b2bc2636b"},"schema_version":"1.0"},"canonical_sha256":"b4c3dd6621663790063cd9327433596398c255e45a3ecdf70d19a7ba60f23337","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:11.937930Z","signature_b64":"SgsqJqRDF9gyLOroTYyEiYdPmqcC4+bFOljK/6AmM5VQxqbJbZMlNPmpp2WBkaF1LUGPHe8mNSRNxQI5gX+EAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b4c3dd6621663790063cd9327433596398c255e45a3ecdf70d19a7ba60f23337","last_reissued_at":"2026-05-17T23:39:11.937158Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:11.937158Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.14117","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":"2026-05-17T23:39:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"I6AAI3d5gab7eibsUnVL9xTFiCZ6/rbMKuS2/L03ZbcRraYV0otuY2gD/ERpzY7/aiZiKfzAcEXSAOuCVpoICw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T01:37:08.683437Z"},"content_sha256":"e5f5f44eb5ed3111428cfe6ccc44a25022377c7510a04b0fb60138aecb1605ac","schema_version":"1.0","event_id":"sha256:e5f5f44eb5ed3111428cfe6ccc44a25022377c7510a04b0fb60138aecb1605ac"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:WTB52ZRBMY3ZABR43EZHIM2ZMO","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Generative Floor Plan Design with LLMs via Reinforcement Learning with Verifiable Rewards","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Fine-tuning an LLM then applying reinforcement learning with verifiable rewards produces floor plans that meet both room connectivity and exact numerical constraints on dimensions and areas.","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Aditya Sharma, Aristides Milios, Christopher Beckham, Christopher Pal, Florian Golemo, Ge Ya Luo, Luis Lara, Zhi Hao Luo","submitted_at":"2026-05-13T21:06:12Z","abstract_excerpt":"An AI system for professional floor plan design must precisely control room dimensions and areas while respecting the desired connectivity between rooms and maintaining functional and aesthetic quality. Existing generative approaches focus primarily on respecting the requested connectivity between rooms, but do not support generating floor plans that respect numerical constraints. We introduce a text-based floor plan generation approach that fine-tunes a large language model (LLM) on real plans and then applies reinforcement learning with verifiable rewards (RLVR) to improve adherence to topol"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Our model generates floor plans that satisfy user-defined connectivity and numerical constraints and outperforms existing methods on Realism, Compatibility, and Diversity metrics. Across all tasks, our approach achieves at least a 94% relative reduction in Compatibility compared with existing methods.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The verifiable rewards and constraint adherence metrics fully capture functional validity of floor plans without overlooking buildability, safety, or aesthetic factors that matter in real professional use.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Fine-tuned LLMs trained with reinforcement learning using verifiable rewards produce floor plans that satisfy connectivity and numerical constraints, outperforming prior methods with at least 94% relative improvement in compatibility.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Fine-tuning an LLM then applying reinforcement learning with verifiable rewards produces floor plans that meet both room connectivity and exact numerical constraints on dimensions and areas.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"8119d677068a7efaaaabb068fe833dcf5cb08b3d30be74446345d01b142abdf1"},"source":{"id":"2605.14117","kind":"arxiv","version":1},"verdict":{"id":"9ca44f5c-62cd-4a42-8f22-7c0f961237e2","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T04:58:11.327114Z","strongest_claim":"Our model generates floor plans that satisfy user-defined connectivity and numerical constraints and outperforms existing methods on Realism, Compatibility, and Diversity metrics. Across all tasks, our approach achieves at least a 94% relative reduction in Compatibility compared with existing methods.","one_line_summary":"Fine-tuned LLMs trained with reinforcement learning using verifiable rewards produce floor plans that satisfy connectivity and numerical constraints, outperforming prior methods with at least 94% relative improvement in compatibility.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The verifiable rewards and constraint adherence metrics fully capture functional validity of floor plans without overlooking buildability, safety, or aesthetic factors that matter in real professional use.","pith_extraction_headline":"Fine-tuning an LLM then applying reinforcement learning with verifiable rewards produces floor plans that meet both room connectivity and exact numerical constraints on dimensions and areas."},"references":{"count":16,"sample":[{"doi":"","year":null,"title":"ArXiv:2506.14702v1","work_id":"eccd4a10-9ccd-4a4a-b664-a6f209abeb79","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Architext: Language-driven generative architecture design.arXiv preprint arXiv:2303.07519. Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al- Dahle, Aiesha ","work_id":"eaa83c63-b0ab-4409-b2c2-0fdeff5b7456","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"The Llama 3 Herd of Models","work_id":"1549a635-88af-4ac1-acfe-51ae7bb53345","ref_index":3,"cited_arxiv_id":"2407.21783","is_internal_anchor":true},{"doi":"","year":null,"title":"LoRA: Low-Rank Adaptation of Large Language Models","work_id":"0426219a-789e-4964-adc8-a04538510818","ref_index":4,"cited_arxiv_id":"2106.09685","is_internal_anchor":true},{"doi":"","year":null,"title":"Efficient Memory Management for Large Language Model Serving with PagedAttention","work_id":"0eb5eca2-2c11-4a77-a25a-18c331a50ed2","ref_index":5,"cited_arxiv_id":"2309.06180","is_internal_anchor":true}],"resolved_work":16,"snapshot_sha256":"89e46dda507419fb0512c95dcbd8bd095bdb2812ae624d2e0fd22790f18cd0cd","internal_anchors":5},"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":"9ca44f5c-62cd-4a42-8f22-7c0f961237e2"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wV6TQ1PY+NDg3z3QcFjfR2GTzY4evTvlF0kV//ZodRU1NWJgwYJ9IjxAaKw9nfBPIcEh0a9niIEFT2B6UwkhBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T01:37:08.684334Z"},"content_sha256":"ea2113dcd525db9c864f2a19916e82abfcd2284d179cee82a156d3fe97383b35","schema_version":"1.0","event_id":"sha256:ea2113dcd525db9c864f2a19916e82abfcd2284d179cee82a156d3fe97383b35"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/WTB52ZRBMY3ZABR43EZHIM2ZMO/bundle.json","state_url":"https://pith.science/pith/WTB52ZRBMY3ZABR43EZHIM2ZMO/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/WTB52ZRBMY3ZABR43EZHIM2ZMO/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":"2026-05-27T01:37:08Z","links":{"resolver":"https://pith.science/pith/WTB52ZRBMY3ZABR43EZHIM2ZMO","bundle":"https://pith.science/pith/WTB52ZRBMY3ZABR43EZHIM2ZMO/bundle.json","state":"https://pith.science/pith/WTB52ZRBMY3ZABR43EZHIM2ZMO/state.json","well_known_bundle":"https://pith.science/.well-known/pith/WTB52ZRBMY3ZABR43EZHIM2ZMO/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:WTB52ZRBMY3ZABR43EZHIM2ZMO","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":"41f82cd9c756bc9a80d2f669439e4d1a14438dcbdc180145e816a33b2bc2636b","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-13T21:06:12Z","title_canon_sha256":"d81df676a297d8a4be1bb03038b110a20ce3989b64b91dc74cb62766389197ee"},"schema_version":"1.0","source":{"id":"2605.14117","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14117","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14117v1","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14117","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"pith_short_12","alias_value":"WTB52ZRBMY3Z","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"WTB52ZRBMY3ZABR4","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"WTB52ZRB","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:ea2113dcd525db9c864f2a19916e82abfcd2284d179cee82a156d3fe97383b35","target":"graph","created_at":"2026-05-17T23:39:11Z","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":"Our model generates floor plans that satisfy user-defined connectivity and numerical constraints and outperforms existing methods on Realism, Compatibility, and Diversity metrics. Across all tasks, our approach achieves at least a 94% relative reduction in Compatibility compared with existing methods."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The verifiable rewards and constraint adherence metrics fully capture functional validity of floor plans without overlooking buildability, safety, or aesthetic factors that matter in real professional use."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Fine-tuned LLMs trained with reinforcement learning using verifiable rewards produce floor plans that satisfy connectivity and numerical constraints, outperforming prior methods with at least 94% relative improvement in compatibility."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Fine-tuning an LLM then applying reinforcement learning with verifiable rewards produces floor plans that meet both room connectivity and exact numerical constraints on dimensions and areas."}],"snapshot_sha256":"8119d677068a7efaaaabb068fe833dcf5cb08b3d30be74446345d01b142abdf1"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"An AI system for professional floor plan design must precisely control room dimensions and areas while respecting the desired connectivity between rooms and maintaining functional and aesthetic quality. Existing generative approaches focus primarily on respecting the requested connectivity between rooms, but do not support generating floor plans that respect numerical constraints. We introduce a text-based floor plan generation approach that fine-tunes a large language model (LLM) on real plans and then applies reinforcement learning with verifiable rewards (RLVR) to improve adherence to topol","authors_text":"Aditya Sharma, Aristides Milios, Christopher Beckham, Christopher Pal, Florian Golemo, Ge Ya Luo, Luis Lara, Zhi Hao Luo","cross_cats":["cs.AI"],"headline":"Fine-tuning an LLM then applying reinforcement learning with verifiable rewards produces floor plans that meet both room connectivity and exact numerical constraints on dimensions and areas.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-13T21:06:12Z","title":"Generative Floor Plan Design with LLMs via Reinforcement Learning with Verifiable Rewards"},"references":{"count":16,"internal_anchors":5,"resolved_work":16,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"ArXiv:2506.14702v1","work_id":"eccd4a10-9ccd-4a4a-b664-a6f209abeb79","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Architext: Language-driven generative architecture design.arXiv preprint arXiv:2303.07519. Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al- Dahle, Aiesha ","work_id":"eaa83c63-b0ab-4409-b2c2-0fdeff5b7456","year":null},{"cited_arxiv_id":"2407.21783","doi":"","is_internal_anchor":true,"ref_index":3,"title":"The Llama 3 Herd of Models","work_id":"1549a635-88af-4ac1-acfe-51ae7bb53345","year":null},{"cited_arxiv_id":"2106.09685","doi":"","is_internal_anchor":true,"ref_index":4,"title":"LoRA: Low-Rank Adaptation of Large Language Models","work_id":"0426219a-789e-4964-adc8-a04538510818","year":null},{"cited_arxiv_id":"2309.06180","doi":"","is_internal_anchor":true,"ref_index":5,"title":"Efficient Memory Management for Large Language Model Serving with PagedAttention","work_id":"0eb5eca2-2c11-4a77-a25a-18c331a50ed2","year":null}],"snapshot_sha256":"89e46dda507419fb0512c95dcbd8bd095bdb2812ae624d2e0fd22790f18cd0cd"},"source":{"id":"2605.14117","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-15T04:58:11.327114Z","id":"9ca44f5c-62cd-4a42-8f22-7c0f961237e2","model_set":{"reader":"grok-4.3"},"one_line_summary":"Fine-tuned LLMs trained with reinforcement learning using verifiable rewards produce floor plans that satisfy connectivity and numerical constraints, outperforming prior methods with at least 94% relative improvement in compatibility.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Fine-tuning an LLM then applying reinforcement learning with verifiable rewards produces floor plans that meet both room connectivity and exact numerical constraints on dimensions and areas.","strongest_claim":"Our model generates floor plans that satisfy user-defined connectivity and numerical constraints and outperforms existing methods on Realism, Compatibility, and Diversity metrics. Across all tasks, our approach achieves at least a 94% relative reduction in Compatibility compared with existing methods.","weakest_assumption":"The verifiable rewards and constraint adherence metrics fully capture functional validity of floor plans without overlooking buildability, safety, or aesthetic factors that matter in real professional use."}},"verdict_id":"9ca44f5c-62cd-4a42-8f22-7c0f961237e2"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:e5f5f44eb5ed3111428cfe6ccc44a25022377c7510a04b0fb60138aecb1605ac","target":"record","created_at":"2026-05-17T23:39:11Z","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":"41f82cd9c756bc9a80d2f669439e4d1a14438dcbdc180145e816a33b2bc2636b","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-13T21:06:12Z","title_canon_sha256":"d81df676a297d8a4be1bb03038b110a20ce3989b64b91dc74cb62766389197ee"},"schema_version":"1.0","source":{"id":"2605.14117","kind":"arxiv","version":1}},"canonical_sha256":"b4c3dd6621663790063cd9327433596398c255e45a3ecdf70d19a7ba60f23337","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b4c3dd6621663790063cd9327433596398c255e45a3ecdf70d19a7ba60f23337","first_computed_at":"2026-05-17T23:39:11.937158Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:11.937158Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"SgsqJqRDF9gyLOroTYyEiYdPmqcC4+bFOljK/6AmM5VQxqbJbZMlNPmpp2WBkaF1LUGPHe8mNSRNxQI5gX+EAQ==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:11.937930Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.14117","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e5f5f44eb5ed3111428cfe6ccc44a25022377c7510a04b0fb60138aecb1605ac","sha256:ea2113dcd525db9c864f2a19916e82abfcd2284d179cee82a156d3fe97383b35"],"state_sha256":"b94c22198f3b07958d436492d4117f803347378c1e80ed09cde105281a352faa"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mRNLsHxyebz1KgrH4CLgc7+P1jfaineR1/dKZOMjd9SdC+UWUfJEYUxBVpjk7C/muKJr9OUwduLZe1bigtOuBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T01:37:08.688615Z","bundle_sha256":"e833bce4ce66d79f021a78209aef163861502c756a9be5f0d64c98419a5cbc18"}}