{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:IUPIMB5C36GEI2RP4E6NZVBSIQ","short_pith_number":"pith:IUPIMB5C","canonical_record":{"source":{"id":"2108.10470","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2021-08-24T01:38:11Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"ce285511ef605af97d467dd54a88a760c46142a1ee80b9a4882a46a3466923f8","abstract_canon_sha256":"83f43b3d5ca4f928fdf785bcf6f28f552bc82a931a0674cf13c20c34e08b0c84"},"schema_version":"1.0"},"canonical_sha256":"451e8607a2df8c446a2fe13cdcd43244206ee6b49aef4cb2a37bb07b63c3b103","source":{"kind":"arxiv","id":"2108.10470","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2108.10470","created_at":"2026-07-05T03:08:56Z"},{"alias_kind":"arxiv_version","alias_value":"2108.10470v2","created_at":"2026-07-05T03:08:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2108.10470","created_at":"2026-07-05T03:08:56Z"},{"alias_kind":"pith_short_12","alias_value":"IUPIMB5C36GE","created_at":"2026-07-05T03:08:56Z"},{"alias_kind":"pith_short_16","alias_value":"IUPIMB5C36GEI2RP","created_at":"2026-07-05T03:08:56Z"},{"alias_kind":"pith_short_8","alias_value":"IUPIMB5C","created_at":"2026-07-05T03:08:56Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:IUPIMB5C36GEI2RP4E6NZVBSIQ","target":"record","payload":{"canonical_record":{"source":{"id":"2108.10470","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2021-08-24T01:38:11Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"ce285511ef605af97d467dd54a88a760c46142a1ee80b9a4882a46a3466923f8","abstract_canon_sha256":"83f43b3d5ca4f928fdf785bcf6f28f552bc82a931a0674cf13c20c34e08b0c84"},"schema_version":"1.0"},"canonical_sha256":"451e8607a2df8c446a2fe13cdcd43244206ee6b49aef4cb2a37bb07b63c3b103","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:08:56.782061Z","signature_b64":"+lGHdxPaG8Q8a4Q/v/ZqkmeLxe4kJ0L0lHknTJqynvHEeC+wD5h5ab9/nWyVFlWRAaprKpW1zY/Y9/euwFetBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"451e8607a2df8c446a2fe13cdcd43244206ee6b49aef4cb2a37bb07b63c3b103","last_reissued_at":"2026-07-05T03:08:56.781576Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:08:56.781576Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2108.10470","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":"2026-07-05T03:08:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cZ5y8uKLv5etk1C3WR3ed5ZoFFw+PxKX9yzL5pjFIpDxLkDHeYFw/u9l4boXRnKKZGBNCiJNs69LqikR7mYtAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T12:23:25.137754Z"},"content_sha256":"f12ef16c1d0a3b241da4e14c8468221967f6b4f6c242c5556e7d86892bd8cfbe","schema_version":"1.0","event_id":"sha256:f12ef16c1d0a3b241da4e14c8468221967f6b4f6c242c5556e7d86892bd8cfbe"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:IUPIMB5C36GEI2RP4E6NZVBSIQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Isaac Gym trains robot policies entirely on one GPU by moving data directly between physics buffers and PyTorch tensors.","cross_cats":["cs.LG"],"primary_cat":"cs.RO","authors_text":"Ankur Handa, Arthur Allshire, David Hoeller, Gavriel State, Kier Storey, Lukasz Wawrzyniak, Michelle Lu, Miles Macklin, Nikita Rudin, Viktor Makoviychuk, Yunrong Guo","submitted_at":"2021-08-24T01:38:11Z","abstract_excerpt":"Isaac Gym offers a high performance learning platform to train policies for wide variety of robotics tasks directly on GPU. Both physics simulation and the neural network policy training reside on GPU and communicate by directly passing data from physics buffers to PyTorch tensors without ever going through any CPU bottlenecks. This leads to blazing fast training times for complex robotics tasks on a single GPU with 2-3 orders of magnitude improvements compared to conventional RL training that uses a CPU based simulator and GPU for neural networks. We host the results and videos at \\url{https:"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Both physics simulation and the neural network policy training reside on GPU and communicate by directly passing data from physics buffers to PyTorch tensors without ever going through any CPU bottlenecks. This leads to blazing fast training times for complex robotics tasks on a single GPU with 2-3 orders of magnitude improvements compared to conventional RL training that uses a CPU based simulator and GPU for neural networks.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The GPU physics engine produces sufficiently accurate and stable results for the target robotics tasks, and that direct GPU buffer sharing introduces no hidden synchronization or numerical artifacts that would invalidate learned policies when transferred to real robots.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Isaac Gym achieves 2-3 orders of magnitude faster robot policy training by keeping physics simulation and PyTorch-based RL entirely on GPU with direct buffer sharing.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Isaac Gym trains robot policies entirely on one GPU by moving data directly between physics buffers and PyTorch tensors.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"8cdb9db38c7ad9a4d245cf54693414c4b7ebdfca1add27f4ae22bbd83d639a25"},"source":{"id":"2108.10470","kind":"arxiv","version":2},"verdict":{"id":"1fe36362-2465-46e0-b3c8-9c1cc79b661a","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-12T21:39:24.424114Z","strongest_claim":"Both physics simulation and the neural network policy training reside on GPU and communicate by directly passing data from physics buffers to PyTorch tensors without ever going through any CPU bottlenecks. This leads to blazing fast training times for complex robotics tasks on a single GPU with 2-3 orders of magnitude improvements compared to conventional RL training that uses a CPU based simulator and GPU for neural networks.","one_line_summary":"Isaac Gym achieves 2-3 orders of magnitude faster robot policy training by keeping physics simulation and PyTorch-based RL entirely on GPU with direct buffer sharing.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The GPU physics engine produces sufficiently accurate and stable results for the target robotics tasks, and that direct GPU buffer sharing introduces no hidden synchronization or numerical artifacts that would invalidate learned policies when transferred to real robots.","pith_extraction_headline":"Isaac Gym trains robot policies entirely on one GPU by moving data directly between physics buffers and PyTorch tensors."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2108.10470/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":39,"sample":[{"doi":"","year":2018,"title":"A general reinforcement learning algorithm that masters chess, shogi, and go through self-play","work_id":"f7c0c123-75f5-4fb3-8e28-2398b6130b83","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2013,"title":"A survey of real-time strategy game ai research and competition in starcraft","work_id":"26f62a38-80dc-4b60-9a7f-91362f83c817","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1912,"title":"Dota 2 with Large Scale Deep Reinforcement Learning","work_id":"b047dc18-e9a3-4d11-8ff6-cd59d41a6357","ref_index":3,"cited_arxiv_id":"1912.06680","is_internal_anchor":true},{"doi":"10.15607/rss.2020.xvi.064","year":2020,"title":"Robotics: Science and Systems (2020) https://doi.org/10.15607/RSS.2020.XVI.064","work_id":"495636fe-f34b-4553-8610-6afb6454ec03","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2018,"title":"Learning Dexterous In-Hand Manipulation","work_id":"389579f0-1ebf-484d-b67d-d8239d049ece","ref_index":5,"cited_arxiv_id":"1808.00177","is_internal_anchor":false}],"resolved_work":39,"snapshot_sha256":"882d0a74652577e716ed5e513f079a8ac5ede5f4441d04f6663eafb6f08abc01","internal_anchors":2},"formal_canon":{"evidence_count":2,"snapshot_sha256":"7723973e3dcf19bd4834b74cd7a660b01fd41bc21c69ce647b28902dfde0d88f"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"1fe36362-2465-46e0-b3c8-9c1cc79b661a"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T03:08:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ygtr68Mx7lnYi9LBfEsWk7Q+lI2pyzsVGUteIs0V8/p0rcQuUjJrgmpq23+pju8g3OiJsJj+P5sChxP+0amMDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T12:23:25.138286Z"},"content_sha256":"56f2c9449c13d4a06d88385df57d2b1739bea1c55e092312b60a900e059d8f44","schema_version":"1.0","event_id":"sha256:56f2c9449c13d4a06d88385df57d2b1739bea1c55e092312b60a900e059d8f44"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/IUPIMB5C36GEI2RP4E6NZVBSIQ/bundle.json","state_url":"https://pith.science/pith/IUPIMB5C36GEI2RP4E6NZVBSIQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/IUPIMB5C36GEI2RP4E6NZVBSIQ/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":"2026-07-07T12:23:25Z","links":{"resolver":"https://pith.science/pith/IUPIMB5C36GEI2RP4E6NZVBSIQ","bundle":"https://pith.science/pith/IUPIMB5C36GEI2RP4E6NZVBSIQ/bundle.json","state":"https://pith.science/pith/IUPIMB5C36GEI2RP4E6NZVBSIQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/IUPIMB5C36GEI2RP4E6NZVBSIQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:IUPIMB5C36GEI2RP4E6NZVBSIQ","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":"83f43b3d5ca4f928fdf785bcf6f28f552bc82a931a0674cf13c20c34e08b0c84","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2021-08-24T01:38:11Z","title_canon_sha256":"ce285511ef605af97d467dd54a88a760c46142a1ee80b9a4882a46a3466923f8"},"schema_version":"1.0","source":{"id":"2108.10470","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2108.10470","created_at":"2026-07-05T03:08:56Z"},{"alias_kind":"arxiv_version","alias_value":"2108.10470v2","created_at":"2026-07-05T03:08:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2108.10470","created_at":"2026-07-05T03:08:56Z"},{"alias_kind":"pith_short_12","alias_value":"IUPIMB5C36GE","created_at":"2026-07-05T03:08:56Z"},{"alias_kind":"pith_short_16","alias_value":"IUPIMB5C36GEI2RP","created_at":"2026-07-05T03:08:56Z"},{"alias_kind":"pith_short_8","alias_value":"IUPIMB5C","created_at":"2026-07-05T03:08:56Z"}],"graph_snapshots":[{"event_id":"sha256:56f2c9449c13d4a06d88385df57d2b1739bea1c55e092312b60a900e059d8f44","target":"graph","created_at":"2026-07-05T03:08:56Z","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":"Both physics simulation and the neural network policy training reside on GPU and communicate by directly passing data from physics buffers to PyTorch tensors without ever going through any CPU bottlenecks. This leads to blazing fast training times for complex robotics tasks on a single GPU with 2-3 orders of magnitude improvements compared to conventional RL training that uses a CPU based simulator and GPU for neural networks."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The GPU physics engine produces sufficiently accurate and stable results for the target robotics tasks, and that direct GPU buffer sharing introduces no hidden synchronization or numerical artifacts that would invalidate learned policies when transferred to real robots."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Isaac Gym achieves 2-3 orders of magnitude faster robot policy training by keeping physics simulation and PyTorch-based RL entirely on GPU with direct buffer sharing."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Isaac Gym trains robot policies entirely on one GPU by moving data directly between physics buffers and PyTorch tensors."}],"snapshot_sha256":"8cdb9db38c7ad9a4d245cf54693414c4b7ebdfca1add27f4ae22bbd83d639a25"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"7723973e3dcf19bd4834b74cd7a660b01fd41bc21c69ce647b28902dfde0d88f"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2108.10470/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Isaac Gym offers a high performance learning platform to train policies for wide variety of robotics tasks directly on GPU. Both physics simulation and the neural network policy training reside on GPU and communicate by directly passing data from physics buffers to PyTorch tensors without ever going through any CPU bottlenecks. This leads to blazing fast training times for complex robotics tasks on a single GPU with 2-3 orders of magnitude improvements compared to conventional RL training that uses a CPU based simulator and GPU for neural networks. We host the results and videos at \\url{https:","authors_text":"Ankur Handa, Arthur Allshire, David Hoeller, Gavriel State, Kier Storey, Lukasz Wawrzyniak, Michelle Lu, Miles Macklin, Nikita Rudin, Viktor Makoviychuk, Yunrong Guo","cross_cats":["cs.LG"],"headline":"Isaac Gym trains robot policies entirely on one GPU by moving data directly between physics buffers and PyTorch tensors.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2021-08-24T01:38:11Z","title":"Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learning"},"references":{"count":39,"internal_anchors":2,"resolved_work":39,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"A general reinforcement learning algorithm that masters chess, shogi, and go through self-play","work_id":"f7c0c123-75f5-4fb3-8e28-2398b6130b83","year":2018},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"A survey of real-time strategy game ai research and competition in starcraft","work_id":"26f62a38-80dc-4b60-9a7f-91362f83c817","year":2013},{"cited_arxiv_id":"1912.06680","doi":"","is_internal_anchor":true,"ref_index":3,"title":"Dota 2 with Large Scale Deep Reinforcement Learning","work_id":"b047dc18-e9a3-4d11-8ff6-cd59d41a6357","year":1912},{"cited_arxiv_id":"","doi":"10.15607/rss.2020.xvi.064","is_internal_anchor":false,"ref_index":4,"title":"Robotics: Science and Systems (2020) https://doi.org/10.15607/RSS.2020.XVI.064","work_id":"495636fe-f34b-4553-8610-6afb6454ec03","year":2020},{"cited_arxiv_id":"1808.00177","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Learning Dexterous In-Hand Manipulation","work_id":"389579f0-1ebf-484d-b67d-d8239d049ece","year":2018}],"snapshot_sha256":"882d0a74652577e716ed5e513f079a8ac5ede5f4441d04f6663eafb6f08abc01"},"source":{"id":"2108.10470","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-12T21:39:24.424114Z","id":"1fe36362-2465-46e0-b3c8-9c1cc79b661a","model_set":{"reader":"grok-4.3"},"one_line_summary":"Isaac Gym achieves 2-3 orders of magnitude faster robot policy training by keeping physics simulation and PyTorch-based RL entirely on GPU with direct buffer sharing.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Isaac Gym trains robot policies entirely on one GPU by moving data directly between physics buffers and PyTorch tensors.","strongest_claim":"Both physics simulation and the neural network policy training reside on GPU and communicate by directly passing data from physics buffers to PyTorch tensors without ever going through any CPU bottlenecks. This leads to blazing fast training times for complex robotics tasks on a single GPU with 2-3 orders of magnitude improvements compared to conventional RL training that uses a CPU based simulator and GPU for neural networks.","weakest_assumption":"The GPU physics engine produces sufficiently accurate and stable results for the target robotics tasks, and that direct GPU buffer sharing introduces no hidden synchronization or numerical artifacts that would invalidate learned policies when transferred to real robots."}},"verdict_id":"1fe36362-2465-46e0-b3c8-9c1cc79b661a"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:f12ef16c1d0a3b241da4e14c8468221967f6b4f6c242c5556e7d86892bd8cfbe","target":"record","created_at":"2026-07-05T03:08:56Z","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":"83f43b3d5ca4f928fdf785bcf6f28f552bc82a931a0674cf13c20c34e08b0c84","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2021-08-24T01:38:11Z","title_canon_sha256":"ce285511ef605af97d467dd54a88a760c46142a1ee80b9a4882a46a3466923f8"},"schema_version":"1.0","source":{"id":"2108.10470","kind":"arxiv","version":2}},"canonical_sha256":"451e8607a2df8c446a2fe13cdcd43244206ee6b49aef4cb2a37bb07b63c3b103","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"451e8607a2df8c446a2fe13cdcd43244206ee6b49aef4cb2a37bb07b63c3b103","first_computed_at":"2026-07-05T03:08:56.781576Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T03:08:56.781576Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"+lGHdxPaG8Q8a4Q/v/ZqkmeLxe4kJ0L0lHknTJqynvHEeC+wD5h5ab9/nWyVFlWRAaprKpW1zY/Y9/euwFetBg==","signature_status":"signed_v1","signed_at":"2026-07-05T03:08:56.782061Z","signed_message":"canonical_sha256_bytes"},"source_id":"2108.10470","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f12ef16c1d0a3b241da4e14c8468221967f6b4f6c242c5556e7d86892bd8cfbe","sha256:56f2c9449c13d4a06d88385df57d2b1739bea1c55e092312b60a900e059d8f44"],"state_sha256":"65f77e3829a4e098b23c71e806667c4f8c38afd4b25c20540eb583847e4fdb4e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Ql4Q/Pn9QL8xcb8QYHec9HXxLvcMTcySzJjmrDroizv+S5RrZ2sQtrHM/aG8JyLjQbmlAbxTveYxR0x2KpXoCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T12:23:25.140685Z","bundle_sha256":"cb48ef0fc3e439bf8a5590fdb0365a025a3525451bb3c601e2ac6193b8d8e716"}}