{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:DG2UPY27XQIOX6MUAXTYMXF6FA","short_pith_number":"pith:DG2UPY27","canonical_record":{"source":{"id":"2605.13209","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2026-05-13T08:58:53Z","cross_cats_sorted":["cs.PF"],"title_canon_sha256":"0bae5c5087a6572d16a9701001eb43c7c952fb0b28e08ba828b989642e7f62b1","abstract_canon_sha256":"17be2c985ccd5ae60ba1a490980aa3b7b15d97b6144f6506d7f078b3e76fe4ca"},"schema_version":"1.0"},"canonical_sha256":"19b547e35fbc10ebf99405e7865cbe28353af6cf8a183e72d3c36d89f4ee8d4b","source":{"kind":"arxiv","id":"2605.13209","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.13209","created_at":"2026-05-18T03:08:48Z"},{"alias_kind":"arxiv_version","alias_value":"2605.13209v1","created_at":"2026-05-18T03:08:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13209","created_at":"2026-05-18T03:08:48Z"},{"alias_kind":"pith_short_12","alias_value":"DG2UPY27XQIO","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"DG2UPY27XQIOX6MU","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"DG2UPY27","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:DG2UPY27XQIOX6MUAXTYMXF6FA","target":"record","payload":{"canonical_record":{"source":{"id":"2605.13209","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2026-05-13T08:58:53Z","cross_cats_sorted":["cs.PF"],"title_canon_sha256":"0bae5c5087a6572d16a9701001eb43c7c952fb0b28e08ba828b989642e7f62b1","abstract_canon_sha256":"17be2c985ccd5ae60ba1a490980aa3b7b15d97b6144f6506d7f078b3e76fe4ca"},"schema_version":"1.0"},"canonical_sha256":"19b547e35fbc10ebf99405e7865cbe28353af6cf8a183e72d3c36d89f4ee8d4b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T03:08:48.562081Z","signature_b64":"D5bxVAYONDTvrKBwC4ZwtVXN9xTST611YO7PjqPMdGT4sqaLAO50/ZFXQnDjBO/lOT4PfH/W69gxvmusE0jrBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"19b547e35fbc10ebf99405e7865cbe28353af6cf8a183e72d3c36d89f4ee8d4b","last_reissued_at":"2026-05-18T03:08:48.561514Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T03:08:48.561514Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.13209","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":"2026-05-18T03:08:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1rb0zoECJXYDgoVhdIyY273ctnqi5Hr+oDPbhHGgT/d+QMBDJPJOkdnDyi+nF9WtpviGTx089j4lTTAP1WeuDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T23:04:56.815827Z"},"content_sha256":"c3d58a828cbd78395a59a60fa6eb6a0266fff0ab251722f4f079342e644fe9c9","schema_version":"1.0","event_id":"sha256:c3d58a828cbd78395a59a60fa6eb6a0266fff0ab251722f4f079342e644fe9c9"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:DG2UPY27XQIOX6MUAXTYMXF6FA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Comparing the Performance of Heterogeneous Conjugate Gradient and Cholesky Solvers on Various Hardware Using SYCL","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Heterogeneous CPU-GPU implementations of CG and Cholesky solvers run up to 32 percent faster than GPU-only versions for large matrices.","cross_cats":["cs.PF"],"primary_cat":"cs.DC","authors_text":"Alexander Strack, Dirk Pfl\\\"uger, Tim Th\\\"uring","submitted_at":"2026-05-13T08:58:53Z","abstract_excerpt":"Many important real-world applications, such as System Identification with Gaussian Processes, involve solving linear systems with symmetric positive-definite matrices. The iterative CG method and direct solvers based on the Cholesky decomposition are two popular methods that can be applied in this case. Since often very large systems have to be solved when dealing with such real-world scenarios, GPUs are commonly used to accelerate the computations. However, homogeneous approaches that only leverage the GPU in the system do not take full advantage of the often powerful CPUs located in modern "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"For large matrices, our heterogeneous implementation is up to 32 percent faster for the CG method and up to 29 percent faster for the Cholesky decomposition compared to the corresponding GPU-only implementations. In addition, for large matrices, our heterogeneous implementation of the Cholesky decomposition can achieve at least 12 percent faster runtimes across several systems with GPUs from NVIDIA, AMD, and Intel.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the overhead of data movement and synchronization between CPU and GPU remains low enough for the heterogeneous schedule to outperform a well-tuned GPU-only kernel on the tested matrix sizes and hardware configurations.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Heterogeneous SYCL-based CG and Cholesky solvers deliver up to 32% and 29% faster runtimes than GPU-only versions for large matrices across multiple GPU vendors.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Heterogeneous CPU-GPU implementations of CG and Cholesky solvers run up to 32 percent faster than GPU-only versions for large matrices.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"e7e66e3df3bb8d737e284f36952933a24731e55cfeaebd95d071baf34bfb84de"},"source":{"id":"2605.13209","kind":"arxiv","version":1},"verdict":{"id":"affc0736-1a83-4832-ad7a-48e69f736f21","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T01:59:46.415354Z","strongest_claim":"For large matrices, our heterogeneous implementation is up to 32 percent faster for the CG method and up to 29 percent faster for the Cholesky decomposition compared to the corresponding GPU-only implementations. In addition, for large matrices, our heterogeneous implementation of the Cholesky decomposition can achieve at least 12 percent faster runtimes across several systems with GPUs from NVIDIA, AMD, and Intel.","one_line_summary":"Heterogeneous SYCL-based CG and Cholesky solvers deliver up to 32% and 29% faster runtimes than GPU-only versions for large matrices across multiple GPU vendors.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the overhead of data movement and synchronization between CPU and GPU remains low enough for the heterogeneous schedule to outperform a well-tuned GPU-only kernel on the tested matrix sizes and hardware configurations.","pith_extraction_headline":"Heterogeneous CPU-GPU implementations of CG and Cholesky solvers run up to 32 percent faster than GPU-only versions for large matrices."},"references":{"count":41,"sample":[{"doi":"10.1177/10943420241261960","year":2024,"title":"Ahmad Abdelfattah, Natalie Beams, Robert Carson, Pieter Ghysels, Tzanio Kolev, Thomas Stitt, Arturo Vargas, Stanimire Tomov, and Jack Dongarra. 2024. MAGMA: Enabling exascale performance with accelera","work_id":"c3081cd5-d3d2-4afe-8a32-8a1830dbfe67","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1109/ispa.2012.16","year":2012,"title":"Pedro Alonso, Manuel F. Dolz, Francisco D. Igual, Rafael Mayo, and Enrique S. Quintana-Ortí. 2012. Reducing Energy Consumption of Dense Linear Alge- bra Operations on Hybrid CPU-GPU Platforms. In2012 ","work_id":"a62ceab8-cc58-41db-8e11-76f760d875ef","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1145/3388333.3388658","year":2020,"title":"Aksel Alpay and Vincent Heuveline. 2020. SYCL beyond OpenCL: The archi- tecture, current state and future direction of hipSYCL. InProceedings of the International Workshop on OpenCL (IWOCL ’20). Assoc","work_id":"4c26aae5-f169-4c16-aa08-d0ecc7452f42","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"AMD. 2022. AMD INSTINCT™MI210 ACCELERATOR. https: //www.amd.com/content/dam/amd/en/documents/instinct-business- docs/product-briefs/instinct-mi210-brochure.pdf","work_id":"aad2b645-c9e1-4f35-8efa-65a106eb5c5d","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"AMD. 2023. AMD EPYC™9004 SERIES PROCESSORS. https://www.amd.com/ content/dam/amd/en/documents/epyc-business-docs/datasheets/amd-epyc- 9004-series-processors-datasheet.pdf","work_id":"ffc6ccbc-fe82-46c5-8c16-5d34abde18ba","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":41,"snapshot_sha256":"ab26a99b94445b2d933d2e7233eac77b294c98eb332893d3dd8f738dfd00fd90","internal_anchors":1},"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":"affc0736-1a83-4832-ad7a-48e69f736f21"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T03:08:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0thquboJJTLps1c3hfZP5/ZZn6PzL+eNY3jGHMMeejk0hu2qUfwkak9NXGDLlaslGM3rTiYxP1fNbN3NZbbqAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T23:04:56.817126Z"},"content_sha256":"c9cec9fbeb1338dabf8a058bbe5ad254c312d11955399544ba3d3e54c6961e42","schema_version":"1.0","event_id":"sha256:c9cec9fbeb1338dabf8a058bbe5ad254c312d11955399544ba3d3e54c6961e42"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/DG2UPY27XQIOX6MUAXTYMXF6FA/bundle.json","state_url":"https://pith.science/pith/DG2UPY27XQIOX6MUAXTYMXF6FA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/DG2UPY27XQIOX6MUAXTYMXF6FA/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":"2026-05-31T23:04:56Z","links":{"resolver":"https://pith.science/pith/DG2UPY27XQIOX6MUAXTYMXF6FA","bundle":"https://pith.science/pith/DG2UPY27XQIOX6MUAXTYMXF6FA/bundle.json","state":"https://pith.science/pith/DG2UPY27XQIOX6MUAXTYMXF6FA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/DG2UPY27XQIOX6MUAXTYMXF6FA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:DG2UPY27XQIOX6MUAXTYMXF6FA","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":"17be2c985ccd5ae60ba1a490980aa3b7b15d97b6144f6506d7f078b3e76fe4ca","cross_cats_sorted":["cs.PF"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2026-05-13T08:58:53Z","title_canon_sha256":"0bae5c5087a6572d16a9701001eb43c7c952fb0b28e08ba828b989642e7f62b1"},"schema_version":"1.0","source":{"id":"2605.13209","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.13209","created_at":"2026-05-18T03:08:48Z"},{"alias_kind":"arxiv_version","alias_value":"2605.13209v1","created_at":"2026-05-18T03:08:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13209","created_at":"2026-05-18T03:08:48Z"},{"alias_kind":"pith_short_12","alias_value":"DG2UPY27XQIO","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"DG2UPY27XQIOX6MU","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"DG2UPY27","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:c9cec9fbeb1338dabf8a058bbe5ad254c312d11955399544ba3d3e54c6961e42","target":"graph","created_at":"2026-05-18T03:08:48Z","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":"For large matrices, our heterogeneous implementation is up to 32 percent faster for the CG method and up to 29 percent faster for the Cholesky decomposition compared to the corresponding GPU-only implementations. In addition, for large matrices, our heterogeneous implementation of the Cholesky decomposition can achieve at least 12 percent faster runtimes across several systems with GPUs from NVIDIA, AMD, and Intel."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the overhead of data movement and synchronization between CPU and GPU remains low enough for the heterogeneous schedule to outperform a well-tuned GPU-only kernel on the tested matrix sizes and hardware configurations."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Heterogeneous SYCL-based CG and Cholesky solvers deliver up to 32% and 29% faster runtimes than GPU-only versions for large matrices across multiple GPU vendors."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Heterogeneous CPU-GPU implementations of CG and Cholesky solvers run up to 32 percent faster than GPU-only versions for large matrices."}],"snapshot_sha256":"e7e66e3df3bb8d737e284f36952933a24731e55cfeaebd95d071baf34bfb84de"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Many important real-world applications, such as System Identification with Gaussian Processes, involve solving linear systems with symmetric positive-definite matrices. The iterative CG method and direct solvers based on the Cholesky decomposition are two popular methods that can be applied in this case. Since often very large systems have to be solved when dealing with such real-world scenarios, GPUs are commonly used to accelerate the computations. However, homogeneous approaches that only leverage the GPU in the system do not take full advantage of the often powerful CPUs located in modern ","authors_text":"Alexander Strack, Dirk Pfl\\\"uger, Tim Th\\\"uring","cross_cats":["cs.PF"],"headline":"Heterogeneous CPU-GPU implementations of CG and Cholesky solvers run up to 32 percent faster than GPU-only versions for large matrices.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2026-05-13T08:58:53Z","title":"Comparing the Performance of Heterogeneous Conjugate Gradient and Cholesky Solvers on Various Hardware Using SYCL"},"references":{"count":41,"internal_anchors":1,"resolved_work":41,"sample":[{"cited_arxiv_id":"","doi":"10.1177/10943420241261960","is_internal_anchor":false,"ref_index":1,"title":"Ahmad Abdelfattah, Natalie Beams, Robert Carson, Pieter Ghysels, Tzanio Kolev, Thomas Stitt, Arturo Vargas, Stanimire Tomov, and Jack Dongarra. 2024. MAGMA: Enabling exascale performance with accelera","work_id":"c3081cd5-d3d2-4afe-8a32-8a1830dbfe67","year":2024},{"cited_arxiv_id":"","doi":"10.1109/ispa.2012.16","is_internal_anchor":false,"ref_index":2,"title":"Pedro Alonso, Manuel F. Dolz, Francisco D. Igual, Rafael Mayo, and Enrique S. Quintana-Ortí. 2012. Reducing Energy Consumption of Dense Linear Alge- bra Operations on Hybrid CPU-GPU Platforms. In2012 ","work_id":"a62ceab8-cc58-41db-8e11-76f760d875ef","year":2012},{"cited_arxiv_id":"","doi":"10.1145/3388333.3388658","is_internal_anchor":false,"ref_index":3,"title":"Aksel Alpay and Vincent Heuveline. 2020. SYCL beyond OpenCL: The archi- tecture, current state and future direction of hipSYCL. InProceedings of the International Workshop on OpenCL (IWOCL ’20). Assoc","work_id":"4c26aae5-f169-4c16-aa08-d0ecc7452f42","year":2020},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"AMD. 2022. AMD INSTINCT™MI210 ACCELERATOR. https: //www.amd.com/content/dam/amd/en/documents/instinct-business- docs/product-briefs/instinct-mi210-brochure.pdf","work_id":"aad2b645-c9e1-4f35-8efa-65a106eb5c5d","year":2022},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"AMD. 2023. AMD EPYC™9004 SERIES PROCESSORS. https://www.amd.com/ content/dam/amd/en/documents/epyc-business-docs/datasheets/amd-epyc- 9004-series-processors-datasheet.pdf","work_id":"ffc6ccbc-fe82-46c5-8c16-5d34abde18ba","year":2023}],"snapshot_sha256":"ab26a99b94445b2d933d2e7233eac77b294c98eb332893d3dd8f738dfd00fd90"},"source":{"id":"2605.13209","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-14T01:59:46.415354Z","id":"affc0736-1a83-4832-ad7a-48e69f736f21","model_set":{"reader":"grok-4.3"},"one_line_summary":"Heterogeneous SYCL-based CG and Cholesky solvers deliver up to 32% and 29% faster runtimes than GPU-only versions for large matrices across multiple GPU vendors.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Heterogeneous CPU-GPU implementations of CG and Cholesky solvers run up to 32 percent faster than GPU-only versions for large matrices.","strongest_claim":"For large matrices, our heterogeneous implementation is up to 32 percent faster for the CG method and up to 29 percent faster for the Cholesky decomposition compared to the corresponding GPU-only implementations. In addition, for large matrices, our heterogeneous implementation of the Cholesky decomposition can achieve at least 12 percent faster runtimes across several systems with GPUs from NVIDIA, AMD, and Intel.","weakest_assumption":"That the overhead of data movement and synchronization between CPU and GPU remains low enough for the heterogeneous schedule to outperform a well-tuned GPU-only kernel on the tested matrix sizes and hardware configurations."}},"verdict_id":"affc0736-1a83-4832-ad7a-48e69f736f21"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:c3d58a828cbd78395a59a60fa6eb6a0266fff0ab251722f4f079342e644fe9c9","target":"record","created_at":"2026-05-18T03:08:48Z","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":"17be2c985ccd5ae60ba1a490980aa3b7b15d97b6144f6506d7f078b3e76fe4ca","cross_cats_sorted":["cs.PF"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2026-05-13T08:58:53Z","title_canon_sha256":"0bae5c5087a6572d16a9701001eb43c7c952fb0b28e08ba828b989642e7f62b1"},"schema_version":"1.0","source":{"id":"2605.13209","kind":"arxiv","version":1}},"canonical_sha256":"19b547e35fbc10ebf99405e7865cbe28353af6cf8a183e72d3c36d89f4ee8d4b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"19b547e35fbc10ebf99405e7865cbe28353af6cf8a183e72d3c36d89f4ee8d4b","first_computed_at":"2026-05-18T03:08:48.561514Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T03:08:48.561514Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"D5bxVAYONDTvrKBwC4ZwtVXN9xTST611YO7PjqPMdGT4sqaLAO50/ZFXQnDjBO/lOT4PfH/W69gxvmusE0jrBA==","signature_status":"signed_v1","signed_at":"2026-05-18T03:08:48.562081Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.13209","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c3d58a828cbd78395a59a60fa6eb6a0266fff0ab251722f4f079342e644fe9c9","sha256:c9cec9fbeb1338dabf8a058bbe5ad254c312d11955399544ba3d3e54c6961e42"],"state_sha256":"61e3b16a9815f7c795ca68595c7edd5b0772509aa8def82ac5ce02974a105b5a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3dT+otb03hL8NOHRLIp1VKYQIPD4cXH7j5PSh7dTv8bXGbR0ONhJi4awdVTKhoV+f5GDOJKHDj9isXffVPmUCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-31T23:04:56.826018Z","bundle_sha256":"b3ccd716c61fc6239b28f842c3e1f2c2da223cda380e7a9444b4c5daf4838f19"}}