{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:HLJ5K5JNQDWILRX5C3G3LXVXIM","short_pith_number":"pith:HLJ5K5JN","canonical_record":{"source":{"id":"2605.06546","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-07T16:41:37Z","cross_cats_sorted":[],"title_canon_sha256":"d367caeff836a702b0ca0de80227376d5f748d2680c32fbdc26840af20b0535f","abstract_canon_sha256":"9d622ab51be8f2caf89fe186f8bda7da0d9721e9a70b5c544347803b9802e14f"},"schema_version":"1.0"},"canonical_sha256":"3ad3d5752d80ec85c6fd16cdb5deb7430c2784b98313981c0cbe11b73894319e","source":{"kind":"arxiv","id":"2605.06546","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.06546","created_at":"2026-05-20T01:05:15Z"},{"alias_kind":"arxiv_version","alias_value":"2605.06546v2","created_at":"2026-05-20T01:05:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.06546","created_at":"2026-05-20T01:05:15Z"},{"alias_kind":"pith_short_12","alias_value":"HLJ5K5JNQDWI","created_at":"2026-05-20T01:05:15Z"},{"alias_kind":"pith_short_16","alias_value":"HLJ5K5JNQDWILRX5","created_at":"2026-05-20T01:05:15Z"},{"alias_kind":"pith_short_8","alias_value":"HLJ5K5JN","created_at":"2026-05-20T01:05:15Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:HLJ5K5JNQDWILRX5C3G3LXVXIM","target":"record","payload":{"canonical_record":{"source":{"id":"2605.06546","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-07T16:41:37Z","cross_cats_sorted":[],"title_canon_sha256":"d367caeff836a702b0ca0de80227376d5f748d2680c32fbdc26840af20b0535f","abstract_canon_sha256":"9d622ab51be8f2caf89fe186f8bda7da0d9721e9a70b5c544347803b9802e14f"},"schema_version":"1.0"},"canonical_sha256":"3ad3d5752d80ec85c6fd16cdb5deb7430c2784b98313981c0cbe11b73894319e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T01:05:15.688427Z","signature_b64":"oh/bxG3Cmew6gV78SzlDmQP05JYRMe/wrSyeji77VDFb3aMFL3sL7p4Awr/wRWDjZLXm8LHsP9EIisNfKpOAAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3ad3d5752d80ec85c6fd16cdb5deb7430c2784b98313981c0cbe11b73894319e","last_reissued_at":"2026-05-20T01:05:15.687504Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T01:05:15.687504Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.06546","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":"2026-05-20T01:05:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ahH2NdLzbR1UhP2865G+FsgkJSQnJI6tpu2Tj/2P5cTbUDKvHyqBp3a5fc1pzjwsRNjPbxBHdNTJ6/kHVNeXBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-05T03:28:32.443168Z"},"content_sha256":"2dd07d58766a6ff089459f0ae7453a64ff92f65c484ded2888ab68e922a9e122","schema_version":"1.0","event_id":"sha256:2dd07d58766a6ff089459f0ae7453a64ff92f65c484ded2888ab68e922a9e122"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:HLJ5K5JNQDWILRX5C3G3LXVXIM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Efficient Pre-Training with Token Superposition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Token superposition in early training reduces large language model pre-training time by up to 2.5 times while matching or exceeding standard results.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bowen Peng, Jeffrey Quesnelle, Th\\'eo Gigant","submitted_at":"2026-05-07T16:41:37Z","abstract_excerpt":"Pre-training of Large Language Models is often prohibitively expensive and inefficient at scale, requiring complex and invasive modifications in order to achieve high data throughput. In this work, we present Token-Superposition Training (TST), a simple drop-in method that significantly improves the data throughput per FLOPs during pre-training without modifying the parallelism, optimizer, tokenizer, data, or model architecture. TST is done in two phases: (i) A highly efficient superposition phase where we combine many contiguous tokens into one bag and train using a multi-hot cross-entropy (M"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"TST consistently outperforms baseline loss and downstream evaluations, and under equal-loss settings, TST yields up to a 2.5x reduction in total pre-training time at the 10B A1B scale.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That any degradation introduced by the superposition phase is fully recoverable in the second phase without extra compute or data, and that the method remains stable and beneficial when scaled beyond the tested 270M–10B range.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Token superposition in an initial training phase followed by recovery allows large language models to reach target loss with substantially less total compute.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Token superposition in early training reduces large language model pre-training time by up to 2.5 times while matching or exceeding standard results.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"a80bd9a6650ea1fb21518dd53771acfdab43b6d46d3df967e6de96d7c976f0a4"},"source":{"id":"2605.06546","kind":"arxiv","version":2},"verdict":{"id":"a4b44701-e136-49ea-9860-a75a603ec4a2","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-08T10:05:13.210849Z","strongest_claim":"TST consistently outperforms baseline loss and downstream evaluations, and under equal-loss settings, TST yields up to a 2.5x reduction in total pre-training time at the 10B A1B scale.","one_line_summary":"Token superposition in an initial training phase followed by recovery allows large language models to reach target loss with substantially less total compute.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That any degradation introduced by the superposition phase is fully recoverable in the second phase without extra compute or data, and that the method remains stable and beneficial when scaled beyond the tested 270M–10B range.","pith_extraction_headline":"Token superposition in early training reduces large language model pre-training time by up to 2.5 times while matching or exceeding standard results."},"integrity":{"clean":false,"summary":{"advisory":0,"critical":1,"by_detector":{"doi_compliance":{"total":1,"advisory":0,"critical":1,"informational":0}},"informational":0},"endpoint":"/pith/2605.06546/integrity.json","findings":[{"note":"Identifier '10.5555/3495724.3497174' is syntactically valid but the DOI registry (doi.org) returned 404, and Crossref / OpenAlex / internal corpus also have no record. The cited work could not be located through any authoritative source.","detector":"doi_compliance","severity":"critical","ref_index":60,"audited_at":"2026-05-19T12:35:38.666874Z","detected_doi":"10.5555/3495724.3497174","finding_type":"unresolvable_identifier","verdict_class":"cross_source","detected_arxiv_id":null}],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-19T18:01:19.568219Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T12:35:38.666874Z","status":"completed","version":"1.0.0","findings_count":1}],"snapshot_sha256":"0e18d1f96cccba09b17ac47037cd0ba9407ee3007461539f1ada02a6543f1968"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"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":"a4b44701-e136-49ea-9860-a75a603ec4a2"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T01:05:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"eTSPhUShIK2+BzrsY3e533wR0vZLjsNeRezVU6lQHhwb2SzozE8EZTSmUaYHxgpCwiRguBjSzcNfmlM9VHEXDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-05T03:28:32.444269Z"},"content_sha256":"9994bcaa1a7cb6f02d70b0a4c570a0a782a59a8bf6cc7b55fbe79c1c90308eb0","schema_version":"1.0","event_id":"sha256:9994bcaa1a7cb6f02d70b0a4c570a0a782a59a8bf6cc7b55fbe79c1c90308eb0"},{"event_type":"integrity_finding","subject_pith_number":"pith:2026:HLJ5K5JNQDWILRX5C3G3LXVXIM","target":"integrity","payload":{"note":"Identifier '10.5555/3495724.3497174' is syntactically valid but the DOI registry (doi.org) returned 404, and Crossref / OpenAlex / internal corpus also have no record. The cited work could not be located through any authoritative source.","snippet":"M. Zaheer, G. Guruganesh, A. Dubey, J. Ainslie, C. Alberti, S. Ontanon, P. Pham, A. Ravula, Q. Wang, L. Yang, and A. Ahmed. Big bird: transformers for longer sequences. InProceedings of the 34th International Conference on Neural Informatio","arxiv_id":"2605.06546","detector":"doi_compliance","evidence":{"doi":"10.5555/3495724.3497174","arxiv_id":null,"ref_index":60,"raw_excerpt":"M. Zaheer, G. Guruganesh, A. Dubey, J. Ainslie, C. Alberti, S. Ontanon, P. Pham, A. Ravula, Q. Wang, L. Yang, and A. Ahmed. Big bird: transformers for longer sequences. InProceedings of the 34th International Conference on Neural Information Processing Systems, NIPS ’20, pages 17283–17297, Red Hook, NY , USA, 2020. Curran Associates Inc. ISBN 978-1-7138- 2954-6. URLhttps://dl.acm.org/doi/10.5555/3","verdict_class":"cross_source","checked_sources":["crossref_by_doi","openalex_by_doi","doi_org_head"]},"severity":"critical","ref_index":60,"audited_at":"2026-05-19T12:35:38.666874Z","event_type":"pith.integrity.v1","detected_doi":"10.5555/3495724.3497174","detector_url":"https://pith.science/pith-integrity-protocol#doi_compliance","external_url":null,"finding_type":"unresolvable_identifier","evidence_hash":"cfe30a81c343e11fb422325b8512a605251dcff0a8fd5449c230b76439f46c36","paper_version":1,"verdict_class":"cross_source","resolved_title":null,"detector_version":"1.0.0","detected_arxiv_id":null,"integrity_event_id":1219,"payload_sha256":"e84625703af9245ec5e2aa22e657e6ecf6bb635bc883ea08312dd270d3d12466","signature_b64":"rpQqNrZcjm4eqymbvksIeWUPxMshzSFG+BL4aFfndoXzNEtXUPRKVDQ8KNG83ubmR8og8AhZWhPj9yvyzmdACA==","signing_key_id":"pith-v1-2026-05"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-19T12:37:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WHmIG553iVjaXvcG7e5yzsVtDUpxGU3+BEBKetoGgJpa5WdNX+qX0CmGNycy7hhZ27eRQPjlHg0zHrhq9GaEBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-05T03:28:32.445684Z"},"content_sha256":"e20cbbbd65edfceb20c4b242eb6bf5c9cec2adf129ac4b61dea863102dd60611","schema_version":"1.0","event_id":"sha256:e20cbbbd65edfceb20c4b242eb6bf5c9cec2adf129ac4b61dea863102dd60611"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HLJ5K5JNQDWILRX5C3G3LXVXIM/bundle.json","state_url":"https://pith.science/pith/HLJ5K5JNQDWILRX5C3G3LXVXIM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HLJ5K5JNQDWILRX5C3G3LXVXIM/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":"2026-06-05T03:28:32Z","links":{"resolver":"https://pith.science/pith/HLJ5K5JNQDWILRX5C3G3LXVXIM","bundle":"https://pith.science/pith/HLJ5K5JNQDWILRX5C3G3LXVXIM/bundle.json","state":"https://pith.science/pith/HLJ5K5JNQDWILRX5C3G3LXVXIM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HLJ5K5JNQDWILRX5C3G3LXVXIM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:HLJ5K5JNQDWILRX5C3G3LXVXIM","merge_version":"pith-open-graph-merge-v1","event_count":3,"valid_event_count":3,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"9d622ab51be8f2caf89fe186f8bda7da0d9721e9a70b5c544347803b9802e14f","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-07T16:41:37Z","title_canon_sha256":"d367caeff836a702b0ca0de80227376d5f748d2680c32fbdc26840af20b0535f"},"schema_version":"1.0","source":{"id":"2605.06546","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.06546","created_at":"2026-05-20T01:05:15Z"},{"alias_kind":"arxiv_version","alias_value":"2605.06546v2","created_at":"2026-05-20T01:05:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.06546","created_at":"2026-05-20T01:05:15Z"},{"alias_kind":"pith_short_12","alias_value":"HLJ5K5JNQDWI","created_at":"2026-05-20T01:05:15Z"},{"alias_kind":"pith_short_16","alias_value":"HLJ5K5JNQDWILRX5","created_at":"2026-05-20T01:05:15Z"},{"alias_kind":"pith_short_8","alias_value":"HLJ5K5JN","created_at":"2026-05-20T01:05:15Z"}],"graph_snapshots":[{"event_id":"sha256:9994bcaa1a7cb6f02d70b0a4c570a0a782a59a8bf6cc7b55fbe79c1c90308eb0","target":"graph","created_at":"2026-05-20T01:05:15Z","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":"TST consistently outperforms baseline loss and downstream evaluations, and under equal-loss settings, TST yields up to a 2.5x reduction in total pre-training time at the 10B A1B scale."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That any degradation introduced by the superposition phase is fully recoverable in the second phase without extra compute or data, and that the method remains stable and beneficial when scaled beyond the tested 270M–10B range."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Token superposition in an initial training phase followed by recovery allows large language models to reach target loss with substantially less total compute."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Token superposition in early training reduces large language model pre-training time by up to 2.5 times while matching or exceeding standard results."}],"snapshot_sha256":"a80bd9a6650ea1fb21518dd53771acfdab43b6d46d3df967e6de96d7c976f0a4"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":false,"detectors_run":[{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T18:01:19.568219Z","status":"completed","version":"1.0.0"},{"findings_count":1,"name":"doi_compliance","ran_at":"2026-05-19T12:35:38.666874Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.06546/integrity.json","findings":[{"audited_at":"2026-05-19T12:35:38.666874Z","detected_arxiv_id":null,"detected_doi":"10.5555/3495724.3497174","detector":"doi_compliance","finding_type":"unresolvable_identifier","note":"Identifier '10.5555/3495724.3497174' is syntactically valid but the DOI registry (doi.org) returned 404, and Crossref / OpenAlex / internal corpus also have no record. The cited work could not be located through any authoritative source.","ref_index":60,"severity":"critical","verdict_class":"cross_source"}],"snapshot_sha256":"0e18d1f96cccba09b17ac47037cd0ba9407ee3007461539f1ada02a6543f1968","summary":{"advisory":0,"by_detector":{"doi_compliance":{"advisory":0,"critical":1,"informational":0,"total":1}},"critical":1,"informational":0}},"paper":{"abstract_excerpt":"Pre-training of Large Language Models is often prohibitively expensive and inefficient at scale, requiring complex and invasive modifications in order to achieve high data throughput. In this work, we present Token-Superposition Training (TST), a simple drop-in method that significantly improves the data throughput per FLOPs during pre-training without modifying the parallelism, optimizer, tokenizer, data, or model architecture. TST is done in two phases: (i) A highly efficient superposition phase where we combine many contiguous tokens into one bag and train using a multi-hot cross-entropy (M","authors_text":"Bowen Peng, Jeffrey Quesnelle, Th\\'eo Gigant","cross_cats":[],"headline":"Token superposition in early training reduces large language model pre-training time by up to 2.5 times while matching or exceeding standard results.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-07T16:41:37Z","title":"Efficient Pre-Training with Token Superposition"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.06546","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-08T10:05:13.210849Z","id":"a4b44701-e136-49ea-9860-a75a603ec4a2","model_set":{"reader":"grok-4.3"},"one_line_summary":"Token superposition in an initial training phase followed by recovery allows large language models to reach target loss with substantially less total compute.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Token superposition in early training reduces large language model pre-training time by up to 2.5 times while matching or exceeding standard results.","strongest_claim":"TST consistently outperforms baseline loss and downstream evaluations, and under equal-loss settings, TST yields up to a 2.5x reduction in total pre-training time at the 10B A1B scale.","weakest_assumption":"That any degradation introduced by the superposition phase is fully recoverable in the second phase without extra compute or data, and that the method remains stable and beneficial when scaled beyond the tested 270M–10B range."}},"verdict_id":"a4b44701-e136-49ea-9860-a75a603ec4a2"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:2dd07d58766a6ff089459f0ae7453a64ff92f65c484ded2888ab68e922a9e122","target":"record","created_at":"2026-05-20T01:05:15Z","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":"9d622ab51be8f2caf89fe186f8bda7da0d9721e9a70b5c544347803b9802e14f","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-07T16:41:37Z","title_canon_sha256":"d367caeff836a702b0ca0de80227376d5f748d2680c32fbdc26840af20b0535f"},"schema_version":"1.0","source":{"id":"2605.06546","kind":"arxiv","version":2}},"canonical_sha256":"3ad3d5752d80ec85c6fd16cdb5deb7430c2784b98313981c0cbe11b73894319e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3ad3d5752d80ec85c6fd16cdb5deb7430c2784b98313981c0cbe11b73894319e","first_computed_at":"2026-05-20T01:05:15.687504Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T01:05:15.687504Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"oh/bxG3Cmew6gV78SzlDmQP05JYRMe/wrSyeji77VDFb3aMFL3sL7p4Awr/wRWDjZLXm8LHsP9EIisNfKpOAAg==","signature_status":"signed_v1","signed_at":"2026-05-20T01:05:15.688427Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.06546","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e20cbbbd65edfceb20c4b242eb6bf5c9cec2adf129ac4b61dea863102dd60611","sha256:2dd07d58766a6ff089459f0ae7453a64ff92f65c484ded2888ab68e922a9e122","sha256:9994bcaa1a7cb6f02d70b0a4c570a0a782a59a8bf6cc7b55fbe79c1c90308eb0"],"state_sha256":"3aa47a24a8c11b8f5f3c91dd86996894e29d44237969552cede2b54d0439cdfc"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mkrNorWH8ZvR1BXQ+4eepphSmfkug6GehOTa7UkoxplWAUUKo37dYlqE1iETYsDMOaCw38TmjCaOwnGgEHXDDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-05T03:28:32.449270Z","bundle_sha256":"532226cfd99cb9ace6ec5e88562f147ecebeb407f656212be9713aae71e59f86"}}