{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:RZXN3SBPEUWBCCLIV6HQA5R36O","short_pith_number":"pith:RZXN3SBP","canonical_record":{"source":{"id":"2605.15237","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AR","submitted_at":"2026-05-14T01:28:16Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"26315e89bb16829fbc8148940bdb7132b537031c4e70d80ab04f9d09f379299c","abstract_canon_sha256":"c7a828d442e7a1ed174b6c43bb29d601bddf53e130c9131f03b375ccc2c8b1c1"},"schema_version":"1.0"},"canonical_sha256":"8e6eddc82f252c110968af8f00763bf39c81c117748e25e6309a42ca17a645c0","source":{"kind":"arxiv","id":"2605.15237","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.15237","created_at":"2026-05-20T00:00:47Z"},{"alias_kind":"arxiv_version","alias_value":"2605.15237v1","created_at":"2026-05-20T00:00:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.15237","created_at":"2026-05-20T00:00:47Z"},{"alias_kind":"pith_short_12","alias_value":"RZXN3SBPEUWB","created_at":"2026-05-20T00:00:47Z"},{"alias_kind":"pith_short_16","alias_value":"RZXN3SBPEUWBCCLI","created_at":"2026-05-20T00:00:47Z"},{"alias_kind":"pith_short_8","alias_value":"RZXN3SBP","created_at":"2026-05-20T00:00:47Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:RZXN3SBPEUWBCCLIV6HQA5R36O","target":"record","payload":{"canonical_record":{"source":{"id":"2605.15237","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AR","submitted_at":"2026-05-14T01:28:16Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"26315e89bb16829fbc8148940bdb7132b537031c4e70d80ab04f9d09f379299c","abstract_canon_sha256":"c7a828d442e7a1ed174b6c43bb29d601bddf53e130c9131f03b375ccc2c8b1c1"},"schema_version":"1.0"},"canonical_sha256":"8e6eddc82f252c110968af8f00763bf39c81c117748e25e6309a42ca17a645c0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:00:47.814795Z","signature_b64":"IOoWiUDujmd3DFFpPHAXUv+rIbJ/W/HTG2ysJ2UvGbcNyg14bZEUnYytSqGet8pCsAGsu0ti9TMZ/48kSDPmBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8e6eddc82f252c110968af8f00763bf39c81c117748e25e6309a42ca17a645c0","last_reissued_at":"2026-05-20T00:00:47.813951Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:00:47.813951Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.15237","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":"2026-05-20T00:00:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mXfGB+tg4ME70alHOAk2kKuAdXQrISFGDOtTGg1e1DvpX3LCDU7HHewRBNsi8P9u57/i2sQGnHvtGHRmPcHNDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T08:07:26.080278Z"},"content_sha256":"1944a3e1f3919e40133daa8410f259f4366243ea404c74919ea80b59534e2dbd","schema_version":"1.0","event_id":"sha256:1944a3e1f3919e40133daa8410f259f4366243ea404c74919ea80b59534e2dbd"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:RZXN3SBPEUWBCCLIV6HQA5R36O","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"A3D: Agentic AI flow for autonomous Accelerator Design","license":"http://creativecommons.org/licenses/by/4.0/","headline":"An agentic AI flow automates the entire hardware accelerator design process for complex applications with no human intervention.","cross_cats":["cs.AI"],"primary_cat":"cs.AR","authors_text":"Abinand Nallathambi, Anand Raghunathan, Christopher Knight, Shantanu Ganguly, Wilfried Haensch","submitted_at":"2026-05-14T01:28:16Z","abstract_excerpt":"Accelerating applications through the design of hardware accelerators can significantly enhance system performance and energy efficiency. Despite advances, such as high-level synthesis (HLS), designing accelerators for complex applications still remains highly labor-intensive, demanding considerable expertise in understanding workloads to be accelerated, hardware design, micro-architecture, and EDA tool usage, posing challenges for application domain experts. Therefore, most accelerator solutions are limited to applications with a regular predictable dataflow. Advances in AI have enabled agent"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Our implementation of A3D, using commercial components like Claude Sonnet 4.5 and the Catapult HLS tool, demonstrates its effectiveness by generating accelerator designs with no human intervention from complex scientific applications like LAMMPS (molecular dynamics simulation) and QMCPACK (quantum chemistry).","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That current LLMs, when partitioned into specialist and verifier agents and augmented with agentic RAG, can reliably perform workload analysis, identify bottlenecks, refactor code for HLS compatibility, and generate correct micro-architectures for complex irregular scientific codes without introducing errors that require human correction.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A3D is an agentic AI system that automates end-to-end hardware accelerator design for complex applications like LAMMPS and QMCPACK with no human intervention.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"An agentic AI flow automates the entire hardware accelerator design process for complex applications with no human intervention.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"d6a6b50892d374b17ade13f6315283b5198fdc1a725404bc4b44d32cd8284881"},"source":{"id":"2605.15237","kind":"arxiv","version":1},"verdict":{"id":"5a8ad6d1-833a-4c66-b486-ef03faffa113","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T16:17:54.867984Z","strongest_claim":"Our implementation of A3D, using commercial components like Claude Sonnet 4.5 and the Catapult HLS tool, demonstrates its effectiveness by generating accelerator designs with no human intervention from complex scientific applications like LAMMPS (molecular dynamics simulation) and QMCPACK (quantum chemistry).","one_line_summary":"A3D is an agentic AI system that automates end-to-end hardware accelerator design for complex applications like LAMMPS and QMCPACK with no human intervention.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That current LLMs, when partitioned into specialist and verifier agents and augmented with agentic RAG, can reliably perform workload analysis, identify bottlenecks, refactor code for HLS compatibility, and generate correct micro-architectures for complex irregular scientific codes without introducing errors that require human correction.","pith_extraction_headline":"An agentic AI flow automates the entire hardware accelerator design process for complex applications with no human intervention."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.15237/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-19T16:31:18.479495Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T16:27:13.075743Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T16:21:55.471466Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T13:33:22.825529Z","status":"skipped","version":"1.0.0","findings_count":0}],"snapshot_sha256":"938803c568a260a03fc028c300a1d0c44e4877549b84f5d08c0a267128e87d14"},"references":{"count":37,"sample":[{"doi":"","year":2024,"title":"Yushi Bai, Xin Lv, Jiajie Zhang, Hongchang Lyu, Jiankai Tang, Zhidian Huang, Zhengxiao Du, Xiao Liu, Aohan Zeng, Lei Hou, et al . 2024. Longbench: A bilingual, multitask benchmark for long context und","work_id":"b6a995a8-e9bf-4490-80bf-543903329f25","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Fengxiang Bie, Yibo Yang, Zhongzhu Zhou, Adam Ghanem, Minjia Zhang, Zhewei Yao, Xiaoxia Wu, Connor Holmes, Pareesa Golnari, David A Clifton, et al. 2024. Renaissance: A survey into ai text-to-image ge","work_id":"798d8de4-4919-4ba7-97ee-15f01c5b0c83","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2016,"title":"Yu-Hsin Chen, Tushar Krishna, Joel S Emer, and Vivienne Sze. 2016. Eyeriss: An energy-efficient reconfigurable accelerator for deep convolutional neural networks.IEEE journal of solid-state circuits52","work_id":"04a04638-2174-4ad2-acdc-91fe778e7098","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2011,"title":"Andrew A Chien, Allan Snavely, and Mark Gahagan. 2011. 10x10: A general- purpose architectural approach to heterogeneity and energy efficiency.Procedia Computer Science4 (2011), 1987–1996","work_id":"909bc27a-2ed2-4215-8e93-f82c78967668","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Luca Collini, Siddharth Garg, and Ramesh Karri. 2025. C2hlsc: Leveraging large language models to bridge the software-to-hardware design gap.ACM Transactions on Design Automation of Electronic Systems","work_id":"29b52b7e-854b-4424-adc0-86248de5e8a0","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":37,"snapshot_sha256":"f0e64fa75fe72f4936955050820291275de86be1d6f6b58ceb8553aa80c9d3be","internal_anchors":7},"formal_canon":{"evidence_count":2,"snapshot_sha256":"c81008e95fd78f4cf30fe1afe67980a281ab9cf7b45649d86b9f96668d5eb058"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"5a8ad6d1-833a-4c66-b486-ef03faffa113"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:00:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5Q9zXgM+i8SppmEMWv6Ybz/rbJryS+Y7Zk2MNamTt6Dnc3oyBUFec7F7122bbQ+i5NLgwQyccWOS4Wfjp3r0Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T08:07:26.081843Z"},"content_sha256":"1a2b41261a6f08fe8b73232d92f1247312babd65763a9165871b8f13b7ea9dc9","schema_version":"1.0","event_id":"sha256:1a2b41261a6f08fe8b73232d92f1247312babd65763a9165871b8f13b7ea9dc9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/RZXN3SBPEUWBCCLIV6HQA5R36O/bundle.json","state_url":"https://pith.science/pith/RZXN3SBPEUWBCCLIV6HQA5R36O/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/RZXN3SBPEUWBCCLIV6HQA5R36O/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":"2026-05-25T08:07:26Z","links":{"resolver":"https://pith.science/pith/RZXN3SBPEUWBCCLIV6HQA5R36O","bundle":"https://pith.science/pith/RZXN3SBPEUWBCCLIV6HQA5R36O/bundle.json","state":"https://pith.science/pith/RZXN3SBPEUWBCCLIV6HQA5R36O/state.json","well_known_bundle":"https://pith.science/.well-known/pith/RZXN3SBPEUWBCCLIV6HQA5R36O/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:RZXN3SBPEUWBCCLIV6HQA5R36O","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":"c7a828d442e7a1ed174b6c43bb29d601bddf53e130c9131f03b375ccc2c8b1c1","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AR","submitted_at":"2026-05-14T01:28:16Z","title_canon_sha256":"26315e89bb16829fbc8148940bdb7132b537031c4e70d80ab04f9d09f379299c"},"schema_version":"1.0","source":{"id":"2605.15237","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.15237","created_at":"2026-05-20T00:00:47Z"},{"alias_kind":"arxiv_version","alias_value":"2605.15237v1","created_at":"2026-05-20T00:00:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.15237","created_at":"2026-05-20T00:00:47Z"},{"alias_kind":"pith_short_12","alias_value":"RZXN3SBPEUWB","created_at":"2026-05-20T00:00:47Z"},{"alias_kind":"pith_short_16","alias_value":"RZXN3SBPEUWBCCLI","created_at":"2026-05-20T00:00:47Z"},{"alias_kind":"pith_short_8","alias_value":"RZXN3SBP","created_at":"2026-05-20T00:00:47Z"}],"graph_snapshots":[{"event_id":"sha256:1a2b41261a6f08fe8b73232d92f1247312babd65763a9165871b8f13b7ea9dc9","target":"graph","created_at":"2026-05-20T00:00:47Z","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 implementation of A3D, using commercial components like Claude Sonnet 4.5 and the Catapult HLS tool, demonstrates its effectiveness by generating accelerator designs with no human intervention from complex scientific applications like LAMMPS (molecular dynamics simulation) and QMCPACK (quantum chemistry)."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That current LLMs, when partitioned into specialist and verifier agents and augmented with agentic RAG, can reliably perform workload analysis, identify bottlenecks, refactor code for HLS compatibility, and generate correct micro-architectures for complex irregular scientific codes without introducing errors that require human correction."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"A3D is an agentic AI system that automates end-to-end hardware accelerator design for complex applications like LAMMPS and QMCPACK with no human intervention."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"An agentic AI flow automates the entire hardware accelerator design process for complex applications with no human intervention."}],"snapshot_sha256":"d6a6b50892d374b17ade13f6315283b5198fdc1a725404bc4b44d32cd8284881"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"c81008e95fd78f4cf30fe1afe67980a281ab9cf7b45649d86b9f96668d5eb058"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T16:31:18.479495Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T16:27:13.075743Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-19T16:21:55.471466Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T13:33:22.825529Z","status":"skipped","version":"1.0.0"}],"endpoint":"/pith/2605.15237/integrity.json","findings":[],"snapshot_sha256":"938803c568a260a03fc028c300a1d0c44e4877549b84f5d08c0a267128e87d14","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Accelerating applications through the design of hardware accelerators can significantly enhance system performance and energy efficiency. Despite advances, such as high-level synthesis (HLS), designing accelerators for complex applications still remains highly labor-intensive, demanding considerable expertise in understanding workloads to be accelerated, hardware design, micro-architecture, and EDA tool usage, posing challenges for application domain experts. Therefore, most accelerator solutions are limited to applications with a regular predictable dataflow. Advances in AI have enabled agent","authors_text":"Abinand Nallathambi, Anand Raghunathan, Christopher Knight, Shantanu Ganguly, Wilfried Haensch","cross_cats":["cs.AI"],"headline":"An agentic AI flow automates the entire hardware accelerator design process for complex applications with no human intervention.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AR","submitted_at":"2026-05-14T01:28:16Z","title":"A3D: Agentic AI flow for autonomous Accelerator Design"},"references":{"count":37,"internal_anchors":7,"resolved_work":37,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Yushi Bai, Xin Lv, Jiajie Zhang, Hongchang Lyu, Jiankai Tang, Zhidian Huang, Zhengxiao Du, Xiao Liu, Aohan Zeng, Lei Hou, et al . 2024. Longbench: A bilingual, multitask benchmark for long context und","work_id":"b6a995a8-e9bf-4490-80bf-543903329f25","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Fengxiang Bie, Yibo Yang, Zhongzhu Zhou, Adam Ghanem, Minjia Zhang, Zhewei Yao, Xiaoxia Wu, Connor Holmes, Pareesa Golnari, David A Clifton, et al. 2024. Renaissance: A survey into ai text-to-image ge","work_id":"798d8de4-4919-4ba7-97ee-15f01c5b0c83","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Yu-Hsin Chen, Tushar Krishna, Joel S Emer, and Vivienne Sze. 2016. Eyeriss: An energy-efficient reconfigurable accelerator for deep convolutional neural networks.IEEE journal of solid-state circuits52","work_id":"04a04638-2174-4ad2-acdc-91fe778e7098","year":2016},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Andrew A Chien, Allan Snavely, and Mark Gahagan. 2011. 10x10: A general- purpose architectural approach to heterogeneity and energy efficiency.Procedia Computer Science4 (2011), 1987–1996","work_id":"909bc27a-2ed2-4215-8e93-f82c78967668","year":2011},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Luca Collini, Siddharth Garg, and Ramesh Karri. 2025. C2hlsc: Leveraging large language models to bridge the software-to-hardware design gap.ACM Transactions on Design Automation of Electronic Systems","work_id":"29b52b7e-854b-4424-adc0-86248de5e8a0","year":2025}],"snapshot_sha256":"f0e64fa75fe72f4936955050820291275de86be1d6f6b58ceb8553aa80c9d3be"},"source":{"id":"2605.15237","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-19T16:17:54.867984Z","id":"5a8ad6d1-833a-4c66-b486-ef03faffa113","model_set":{"reader":"grok-4.3"},"one_line_summary":"A3D is an agentic AI system that automates end-to-end hardware accelerator design for complex applications like LAMMPS and QMCPACK with no human intervention.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"An agentic AI flow automates the entire hardware accelerator design process for complex applications with no human intervention.","strongest_claim":"Our implementation of A3D, using commercial components like Claude Sonnet 4.5 and the Catapult HLS tool, demonstrates its effectiveness by generating accelerator designs with no human intervention from complex scientific applications like LAMMPS (molecular dynamics simulation) and QMCPACK (quantum chemistry).","weakest_assumption":"That current LLMs, when partitioned into specialist and verifier agents and augmented with agentic RAG, can reliably perform workload analysis, identify bottlenecks, refactor code for HLS compatibility, and generate correct micro-architectures for complex irregular scientific codes without introducing errors that require human correction."}},"verdict_id":"5a8ad6d1-833a-4c66-b486-ef03faffa113"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:1944a3e1f3919e40133daa8410f259f4366243ea404c74919ea80b59534e2dbd","target":"record","created_at":"2026-05-20T00:00:47Z","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":"c7a828d442e7a1ed174b6c43bb29d601bddf53e130c9131f03b375ccc2c8b1c1","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AR","submitted_at":"2026-05-14T01:28:16Z","title_canon_sha256":"26315e89bb16829fbc8148940bdb7132b537031c4e70d80ab04f9d09f379299c"},"schema_version":"1.0","source":{"id":"2605.15237","kind":"arxiv","version":1}},"canonical_sha256":"8e6eddc82f252c110968af8f00763bf39c81c117748e25e6309a42ca17a645c0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8e6eddc82f252c110968af8f00763bf39c81c117748e25e6309a42ca17a645c0","first_computed_at":"2026-05-20T00:00:47.813951Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:00:47.813951Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"IOoWiUDujmd3DFFpPHAXUv+rIbJ/W/HTG2ysJ2UvGbcNyg14bZEUnYytSqGet8pCsAGsu0ti9TMZ/48kSDPmBA==","signature_status":"signed_v1","signed_at":"2026-05-20T00:00:47.814795Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.15237","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1944a3e1f3919e40133daa8410f259f4366243ea404c74919ea80b59534e2dbd","sha256:1a2b41261a6f08fe8b73232d92f1247312babd65763a9165871b8f13b7ea9dc9"],"state_sha256":"01a21c3346cd3e463e18166113acfddb758ab3491fe1c5175e94c2e1a910de08"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"irhsFBjtNIP7dWrkU8XfvULXs79RP6be8pxi0do7DRUIo06tp4BXHsLByfF2sZ+C/QdoPqyeYOE0D7oumZCfDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-25T08:07:26.088098Z","bundle_sha256":"093830f1e8a612e241e4682a0110d49c8dc0a76849e5f1da80a2ed87dcc5a323"}}