{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:PYE5AVJQ7LLVLKZHZ3CL2Q4JIZ","short_pith_number":"pith:PYE5AVJQ","canonical_record":{"source":{"id":"2605.15394","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-14T20:27:32Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"8921e762b44b31f71bf28a3d2f0d0d6aa84b965a52da3e407da36f114436b3b8","abstract_canon_sha256":"3de13ac3eea96cb0c4d0c8c3dc748ab683b560aa36b140508fb53cfc0630ac65"},"schema_version":"1.0"},"canonical_sha256":"7e09d05530fad755ab27cec4bd43894671e5ec255d7498f107a4968e91815db3","source":{"kind":"arxiv","id":"2605.15394","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.15394","created_at":"2026-05-20T00:00:56Z"},{"alias_kind":"arxiv_version","alias_value":"2605.15394v1","created_at":"2026-05-20T00:00:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.15394","created_at":"2026-05-20T00:00:56Z"},{"alias_kind":"pith_short_12","alias_value":"PYE5AVJQ7LLV","created_at":"2026-05-20T00:00:56Z"},{"alias_kind":"pith_short_16","alias_value":"PYE5AVJQ7LLVLKZH","created_at":"2026-05-20T00:00:56Z"},{"alias_kind":"pith_short_8","alias_value":"PYE5AVJQ","created_at":"2026-05-20T00:00:56Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:PYE5AVJQ7LLVLKZHZ3CL2Q4JIZ","target":"record","payload":{"canonical_record":{"source":{"id":"2605.15394","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-14T20:27:32Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"8921e762b44b31f71bf28a3d2f0d0d6aa84b965a52da3e407da36f114436b3b8","abstract_canon_sha256":"3de13ac3eea96cb0c4d0c8c3dc748ab683b560aa36b140508fb53cfc0630ac65"},"schema_version":"1.0"},"canonical_sha256":"7e09d05530fad755ab27cec4bd43894671e5ec255d7498f107a4968e91815db3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:00:56.350951Z","signature_b64":"Yi1wFMTixM4P5dkdV87jn39/TfXgJUn0j4AnltuyA5bypPd28l9yhoLx0OJKo9LVC2/tuYPKF28livQpuMbEAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7e09d05530fad755ab27cec4bd43894671e5ec255d7498f107a4968e91815db3","last_reissued_at":"2026-05-20T00:00:56.350056Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:00:56.350056Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.15394","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":"2026-05-20T00:00:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3LDuk7yT/UyFeCchKEldDbBCr0IdzQX2UyWJjxCSHjkAU56vqgPiN1wr5d8ptCH/1UuTMEWcjiCJVbaQg2h7Dg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-24T17:48:37.964895Z"},"content_sha256":"59df235e07115d8eb3f566f5e49e31489c114340274766deaaf19e148f1daf3e","schema_version":"1.0","event_id":"sha256:59df235e07115d8eb3f566f5e49e31489c114340274766deaaf19e148f1daf3e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:PYE5AVJQ7LLVLKZHZ3CL2Q4JIZ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Representation Without Reward: A JEPA Audit for LLM Fine-Tuning","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"JEPA-style auxiliaries change LLM hidden-state geometry but leave task accuracy unchanged on language-to-regex generation","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Biswa Sengupta","submitted_at":"2026-05-14T20:27:32Z","abstract_excerpt":"Joint-embedding predictive architectures (JEPAs) propose that a model should learn more useful abstractions when trained to predict latent representations rather than observed outputs. For autoregressive language-model fine-tuning the principle entails a stricter requirement: the induced hidden-state geometry must reach the language-model head \\emph{and} improve the decoded task metric. We test that requirement under a fixed Llama-3.2-1B-Instruct LoRA harness on natural-language-to-regex generation, comparing twenty-two training-time auxiliaries across trajectory-shape regularisation, distribu"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Hidden-state representation work and decoded-task accuracy are therefore weakly coupled in this regime; we accordingly reframe LLM-domain JEPA evaluation as a coupling problem.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The natural-language-to-regex generation task with exact-match metric is sufficiently representative that a null result on it generalizes to the broader claim of weak coupling between hidden geometry and task signal in LLM fine-tuning.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"An empirical audit of 22 JEPA-style training auxiliaries on Llama-3.2-1B fine-tuning for regex generation finds no statistically significant task improvement after multiple-testing correction, even when auxiliaries visibly alter hidden-state geometry.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"JEPA-style auxiliaries change LLM hidden-state geometry but leave task accuracy unchanged on language-to-regex generation","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"45a132332802796c56a5c9247ecd38d05a94c0c9bd9b694f32be1d228460b9f2"},"source":{"id":"2605.15394","kind":"arxiv","version":1},"verdict":{"id":"0d6c2394-c5b3-423d-919b-058f03351c34","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T16:23:48.105202Z","strongest_claim":"Hidden-state representation work and decoded-task accuracy are therefore weakly coupled in this regime; we accordingly reframe LLM-domain JEPA evaluation as a coupling problem.","one_line_summary":"An empirical audit of 22 JEPA-style training auxiliaries on Llama-3.2-1B fine-tuning for regex generation finds no statistically significant task improvement after multiple-testing correction, even when auxiliaries visibly alter hidden-state geometry.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The natural-language-to-regex generation task with exact-match metric is sufficiently representative that a null result on it generalizes to the broader claim of weak coupling between hidden geometry and task signal in LLM fine-tuning.","pith_extraction_headline":"JEPA-style auxiliaries change LLM hidden-state geometry but leave task accuracy unchanged on language-to-regex generation"},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.15394/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_compliance","ran_at":"2026-05-19T16:35:40.350006Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T16:31:18.273113Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"cited_work_retraction","ran_at":"2026-05-19T16:26:23.648026Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"citation_quote_validity","ran_at":"2026-05-19T15:50:56.442768Z","status":"skipped","version":"0.1.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T14:21:54.166486Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T13:33:22.720708Z","status":"skipped","version":"1.0.0","findings_count":0}],"snapshot_sha256":"c5f36aad65e4137e6f411719cd53ddd55ab35441c106cc8541396cc3c23ead59"},"references":{"count":29,"sample":[{"doi":"","year":1990,"title":"Making the world differentiable: On using self- supervised fully recurrent neural networks for dynamic reinforcement learning and planning in non-stationary environments,","work_id":"45eeca5f-588a-4a31-a19f-358b9256f222","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"A path towards autonomous machine intelligence, ver- sion 0.9.2,","work_id":"c01f7090-2ce8-48c3-a5f9-f404fa23863e","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1991,"title":"Curious model-building control systems,","work_id":"478db2bc-2da1-4b08-8d18-a7ea7dbcfb7c","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Self-supervised learning from images with a joint-embedding predictive architecture.arXiv preprint arXiv:2301.08243","work_id":"ca0a2fcc-b704-4e8a-8977-b8bcba691059","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2026,"title":"arXiv preprint arXiv:2603.14482 (2026)","work_id":"3da7cd7a-c3f3-4e48-9704-4320e54aec60","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":29,"snapshot_sha256":"d4696625df2127a4403f12a61b9e35d3aeb4a95b0049e48ba7f0735147fee475","internal_anchors":6},"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":"0d6c2394-c5b3-423d-919b-058f03351c34"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:00:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5xH1h4x5FugXm1x8QaUEXqqiBfjPmDkHqUN+1cKBepx0dyxzs32quomEqFMDA8cwDfxx7jmRhzqZKAN3w169Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-24T17:48:37.965726Z"},"content_sha256":"a80c7e57ec12d2ce163c1c9b1bdd8d5decb5e5f4d829896ef4feac2986eecd43","schema_version":"1.0","event_id":"sha256:a80c7e57ec12d2ce163c1c9b1bdd8d5decb5e5f4d829896ef4feac2986eecd43"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/PYE5AVJQ7LLVLKZHZ3CL2Q4JIZ/bundle.json","state_url":"https://pith.science/pith/PYE5AVJQ7LLVLKZHZ3CL2Q4JIZ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/PYE5AVJQ7LLVLKZHZ3CL2Q4JIZ/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":"2026-05-24T17:48:37Z","links":{"resolver":"https://pith.science/pith/PYE5AVJQ7LLVLKZHZ3CL2Q4JIZ","bundle":"https://pith.science/pith/PYE5AVJQ7LLVLKZHZ3CL2Q4JIZ/bundle.json","state":"https://pith.science/pith/PYE5AVJQ7LLVLKZHZ3CL2Q4JIZ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/PYE5AVJQ7LLVLKZHZ3CL2Q4JIZ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:PYE5AVJQ7LLVLKZHZ3CL2Q4JIZ","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":"3de13ac3eea96cb0c4d0c8c3dc748ab683b560aa36b140508fb53cfc0630ac65","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-14T20:27:32Z","title_canon_sha256":"8921e762b44b31f71bf28a3d2f0d0d6aa84b965a52da3e407da36f114436b3b8"},"schema_version":"1.0","source":{"id":"2605.15394","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.15394","created_at":"2026-05-20T00:00:56Z"},{"alias_kind":"arxiv_version","alias_value":"2605.15394v1","created_at":"2026-05-20T00:00:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.15394","created_at":"2026-05-20T00:00:56Z"},{"alias_kind":"pith_short_12","alias_value":"PYE5AVJQ7LLV","created_at":"2026-05-20T00:00:56Z"},{"alias_kind":"pith_short_16","alias_value":"PYE5AVJQ7LLVLKZH","created_at":"2026-05-20T00:00:56Z"},{"alias_kind":"pith_short_8","alias_value":"PYE5AVJQ","created_at":"2026-05-20T00:00:56Z"}],"graph_snapshots":[{"event_id":"sha256:a80c7e57ec12d2ce163c1c9b1bdd8d5decb5e5f4d829896ef4feac2986eecd43","target":"graph","created_at":"2026-05-20T00:00:56Z","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":"Hidden-state representation work and decoded-task accuracy are therefore weakly coupled in this regime; we accordingly reframe LLM-domain JEPA evaluation as a coupling problem."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The natural-language-to-regex generation task with exact-match metric is sufficiently representative that a null result on it generalizes to the broader claim of weak coupling between hidden geometry and task signal in LLM fine-tuning."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"An empirical audit of 22 JEPA-style training auxiliaries on Llama-3.2-1B fine-tuning for regex generation finds no statistically significant task improvement after multiple-testing correction, even when auxiliaries visibly alter hidden-state geometry."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"JEPA-style auxiliaries change LLM hidden-state geometry but leave task accuracy unchanged on language-to-regex generation"}],"snapshot_sha256":"45a132332802796c56a5c9247ecd38d05a94c0c9bd9b694f32be1d228460b9f2"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T16:35:40.350006Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T16:31:18.273113Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"cited_work_retraction","ran_at":"2026-05-19T16:26:23.648026Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"citation_quote_validity","ran_at":"2026-05-19T15:50:56.442768Z","status":"skipped","version":"0.1.0"},{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-19T14:21:54.166486Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T13:33:22.720708Z","status":"skipped","version":"1.0.0"}],"endpoint":"/pith/2605.15394/integrity.json","findings":[],"snapshot_sha256":"c5f36aad65e4137e6f411719cd53ddd55ab35441c106cc8541396cc3c23ead59","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Joint-embedding predictive architectures (JEPAs) propose that a model should learn more useful abstractions when trained to predict latent representations rather than observed outputs. For autoregressive language-model fine-tuning the principle entails a stricter requirement: the induced hidden-state geometry must reach the language-model head \\emph{and} improve the decoded task metric. We test that requirement under a fixed Llama-3.2-1B-Instruct LoRA harness on natural-language-to-regex generation, comparing twenty-two training-time auxiliaries across trajectory-shape regularisation, distribu","authors_text":"Biswa Sengupta","cross_cats":["cs.AI","stat.ML"],"headline":"JEPA-style auxiliaries change LLM hidden-state geometry but leave task accuracy unchanged on language-to-regex generation","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-14T20:27:32Z","title":"Representation Without Reward: A JEPA Audit for LLM Fine-Tuning"},"references":{"count":29,"internal_anchors":6,"resolved_work":29,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Making the world differentiable: On using self- supervised fully recurrent neural networks for dynamic reinforcement learning and planning in non-stationary environments,","work_id":"45eeca5f-588a-4a31-a19f-358b9256f222","year":1990},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"A path towards autonomous machine intelligence, ver- sion 0.9.2,","work_id":"c01f7090-2ce8-48c3-a5f9-f404fa23863e","year":2022},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Curious model-building control systems,","work_id":"478db2bc-2da1-4b08-8d18-a7ea7dbcfb7c","year":1991},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Self-supervised learning from images with a joint-embedding predictive architecture.arXiv preprint arXiv:2301.08243","work_id":"ca0a2fcc-b704-4e8a-8977-b8bcba691059","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"arXiv preprint arXiv:2603.14482 (2026)","work_id":"3da7cd7a-c3f3-4e48-9704-4320e54aec60","year":2026}],"snapshot_sha256":"d4696625df2127a4403f12a61b9e35d3aeb4a95b0049e48ba7f0735147fee475"},"source":{"id":"2605.15394","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-19T16:23:48.105202Z","id":"0d6c2394-c5b3-423d-919b-058f03351c34","model_set":{"reader":"grok-4.3"},"one_line_summary":"An empirical audit of 22 JEPA-style training auxiliaries on Llama-3.2-1B fine-tuning for regex generation finds no statistically significant task improvement after multiple-testing correction, even when auxiliaries visibly alter hidden-state geometry.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"JEPA-style auxiliaries change LLM hidden-state geometry but leave task accuracy unchanged on language-to-regex generation","strongest_claim":"Hidden-state representation work and decoded-task accuracy are therefore weakly coupled in this regime; we accordingly reframe LLM-domain JEPA evaluation as a coupling problem.","weakest_assumption":"The natural-language-to-regex generation task with exact-match metric is sufficiently representative that a null result on it generalizes to the broader claim of weak coupling between hidden geometry and task signal in LLM fine-tuning."}},"verdict_id":"0d6c2394-c5b3-423d-919b-058f03351c34"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:59df235e07115d8eb3f566f5e49e31489c114340274766deaaf19e148f1daf3e","target":"record","created_at":"2026-05-20T00:00:56Z","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":"3de13ac3eea96cb0c4d0c8c3dc748ab683b560aa36b140508fb53cfc0630ac65","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-14T20:27:32Z","title_canon_sha256":"8921e762b44b31f71bf28a3d2f0d0d6aa84b965a52da3e407da36f114436b3b8"},"schema_version":"1.0","source":{"id":"2605.15394","kind":"arxiv","version":1}},"canonical_sha256":"7e09d05530fad755ab27cec4bd43894671e5ec255d7498f107a4968e91815db3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7e09d05530fad755ab27cec4bd43894671e5ec255d7498f107a4968e91815db3","first_computed_at":"2026-05-20T00:00:56.350056Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:00:56.350056Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Yi1wFMTixM4P5dkdV87jn39/TfXgJUn0j4AnltuyA5bypPd28l9yhoLx0OJKo9LVC2/tuYPKF28livQpuMbEAQ==","signature_status":"signed_v1","signed_at":"2026-05-20T00:00:56.350951Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.15394","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:59df235e07115d8eb3f566f5e49e31489c114340274766deaaf19e148f1daf3e","sha256:a80c7e57ec12d2ce163c1c9b1bdd8d5decb5e5f4d829896ef4feac2986eecd43"],"state_sha256":"733cd399b0d545a90224d2472d506d09340009f64d74767d51359dd762a2b8ba"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2xMyStv/lWysCFgJ0I1b568935zgOn17R6IhmM+zua8+gjeYHbk/UGpIbFAzujjGtL1lAf9kXOkIgKSZZ80ICA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-24T17:48:37.970009Z","bundle_sha256":"5471fca119c77f514f7c72491809e5336def5315783fdfb99f9c2d08956357a8"}}