{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:AIJZ3GNRPTMF4KEMJ6MQOQY2HT","short_pith_number":"pith:AIJZ3GNR","canonical_record":{"source":{"id":"2606.01008","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-05-31T04:51:27Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"74f5c59d15094843096db894880d258c3bb69049cbc484680a7e16aedca3640a","abstract_canon_sha256":"1db95b934bf8f26e3de29c6ae3cd2677f7b2784ea2c01ca9df0182cfa4677cea"},"schema_version":"1.0"},"canonical_sha256":"02139d99b17cd85e288c4f9907431a3ce3ed0b14cdb0f6a4c50f05a47aa3d68b","source":{"kind":"arxiv","id":"2606.01008","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.01008","created_at":"2026-06-02T01:04:18Z"},{"alias_kind":"arxiv_version","alias_value":"2606.01008v1","created_at":"2026-06-02T01:04:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.01008","created_at":"2026-06-02T01:04:18Z"},{"alias_kind":"pith_short_12","alias_value":"AIJZ3GNRPTMF","created_at":"2026-06-02T01:04:18Z"},{"alias_kind":"pith_short_16","alias_value":"AIJZ3GNRPTMF4KEM","created_at":"2026-06-02T01:04:18Z"},{"alias_kind":"pith_short_8","alias_value":"AIJZ3GNR","created_at":"2026-06-02T01:04:18Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:AIJZ3GNRPTMF4KEMJ6MQOQY2HT","target":"record","payload":{"canonical_record":{"source":{"id":"2606.01008","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-05-31T04:51:27Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"74f5c59d15094843096db894880d258c3bb69049cbc484680a7e16aedca3640a","abstract_canon_sha256":"1db95b934bf8f26e3de29c6ae3cd2677f7b2784ea2c01ca9df0182cfa4677cea"},"schema_version":"1.0"},"canonical_sha256":"02139d99b17cd85e288c4f9907431a3ce3ed0b14cdb0f6a4c50f05a47aa3d68b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T01:04:18.226086Z","signature_b64":"fACuHF5Rj5hgIFTmKFuwhw2s9fhgQzZgy9UkFEPYcrtSx/3lnNgt3OKuMdrMy39EmNAOsMpBv+csx93sRB1xDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"02139d99b17cd85e288c4f9907431a3ce3ed0b14cdb0f6a4c50f05a47aa3d68b","last_reissued_at":"2026-06-02T01:04:18.225456Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T01:04:18.225456Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.01008","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":"2026-06-02T01:04:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9U02N62fWWZwu/JxQZbcLJcLqQNzWCRAYNS+2uQBp2cRU+uS9Dqgy5ZZTDK0Mm3iZAfs9nVo10Af1q4vq6NDCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-29T18:26:00.340575Z"},"content_sha256":"cca8a1bdc3680916a779b9d48926a1806220aa0f4b68c5cbceaa4d985742449d","schema_version":"1.0","event_id":"sha256:cca8a1bdc3680916a779b9d48926a1806220aa0f4b68c5cbceaa4d985742449d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:AIJZ3GNRPTMF4KEMJ6MQOQY2HT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"FVSpec: Real-World Property-Based Tests as Lean Challenges","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SE","authors_text":"Hazel Shackleton, Max von Hippel, Mike Dodds, Quinn Dougherty","submitted_at":"2026-05-31T04:51:27Z","abstract_excerpt":"We present a benchmark for evaluating AI models and agents on real-world formal software verification tasks. We first scrape 11,039 property-based tests (PBTs) from real-world Python repositories, then automatically translate 2,772 of them (25%) into 9,415 Lean 4 specifications with sorry placeholders (about 3 formalizations/PBT; we retain multiple attempts when none dominates on quality metrics). Translating PBTs into Lean specifications is challenging: it requires modeling Python semantics in Lean, inferring the logical property encoded in an imperative PBT, and handling the inherent difficu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.01008","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2606.01008/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"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":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-02T01:04:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+/hW/7qV/CRG0fsNZfjxRbkFbpZiVCudnpIEzXZZL/CwvSuXkefep+D8+e6eslLv1eejjUl0jHNEYHstFzLyCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-29T18:26:00.340949Z"},"content_sha256":"89395b16ffa79400b4c934572fcb478122f84feb95cf53e26df79c5059204095","schema_version":"1.0","event_id":"sha256:89395b16ffa79400b4c934572fcb478122f84feb95cf53e26df79c5059204095"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/AIJZ3GNRPTMF4KEMJ6MQOQY2HT/bundle.json","state_url":"https://pith.science/pith/AIJZ3GNRPTMF4KEMJ6MQOQY2HT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/AIJZ3GNRPTMF4KEMJ6MQOQY2HT/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":"2026-06-29T18:26:00Z","links":{"resolver":"https://pith.science/pith/AIJZ3GNRPTMF4KEMJ6MQOQY2HT","bundle":"https://pith.science/pith/AIJZ3GNRPTMF4KEMJ6MQOQY2HT/bundle.json","state":"https://pith.science/pith/AIJZ3GNRPTMF4KEMJ6MQOQY2HT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/AIJZ3GNRPTMF4KEMJ6MQOQY2HT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:AIJZ3GNRPTMF4KEMJ6MQOQY2HT","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":"1db95b934bf8f26e3de29c6ae3cd2677f7b2784ea2c01ca9df0182cfa4677cea","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-05-31T04:51:27Z","title_canon_sha256":"74f5c59d15094843096db894880d258c3bb69049cbc484680a7e16aedca3640a"},"schema_version":"1.0","source":{"id":"2606.01008","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.01008","created_at":"2026-06-02T01:04:18Z"},{"alias_kind":"arxiv_version","alias_value":"2606.01008v1","created_at":"2026-06-02T01:04:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.01008","created_at":"2026-06-02T01:04:18Z"},{"alias_kind":"pith_short_12","alias_value":"AIJZ3GNRPTMF","created_at":"2026-06-02T01:04:18Z"},{"alias_kind":"pith_short_16","alias_value":"AIJZ3GNRPTMF4KEM","created_at":"2026-06-02T01:04:18Z"},{"alias_kind":"pith_short_8","alias_value":"AIJZ3GNR","created_at":"2026-06-02T01:04:18Z"}],"graph_snapshots":[{"event_id":"sha256:89395b16ffa79400b4c934572fcb478122f84feb95cf53e26df79c5059204095","target":"graph","created_at":"2026-06-02T01:04:18Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2606.01008/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We present a benchmark for evaluating AI models and agents on real-world formal software verification tasks. We first scrape 11,039 property-based tests (PBTs) from real-world Python repositories, then automatically translate 2,772 of them (25%) into 9,415 Lean 4 specifications with sorry placeholders (about 3 formalizations/PBT; we retain multiple attempts when none dominates on quality metrics). Translating PBTs into Lean specifications is challenging: it requires modeling Python semantics in Lean, inferring the logical property encoded in an imperative PBT, and handling the inherent difficu","authors_text":"Hazel Shackleton, Max von Hippel, Mike Dodds, Quinn Dougherty","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-05-31T04:51:27Z","title":"FVSpec: Real-World Property-Based Tests as Lean Challenges"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.01008","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:cca8a1bdc3680916a779b9d48926a1806220aa0f4b68c5cbceaa4d985742449d","target":"record","created_at":"2026-06-02T01:04:18Z","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":"1db95b934bf8f26e3de29c6ae3cd2677f7b2784ea2c01ca9df0182cfa4677cea","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-05-31T04:51:27Z","title_canon_sha256":"74f5c59d15094843096db894880d258c3bb69049cbc484680a7e16aedca3640a"},"schema_version":"1.0","source":{"id":"2606.01008","kind":"arxiv","version":1}},"canonical_sha256":"02139d99b17cd85e288c4f9907431a3ce3ed0b14cdb0f6a4c50f05a47aa3d68b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"02139d99b17cd85e288c4f9907431a3ce3ed0b14cdb0f6a4c50f05a47aa3d68b","first_computed_at":"2026-06-02T01:04:18.225456Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T01:04:18.225456Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fACuHF5Rj5hgIFTmKFuwhw2s9fhgQzZgy9UkFEPYcrtSx/3lnNgt3OKuMdrMy39EmNAOsMpBv+csx93sRB1xDg==","signature_status":"signed_v1","signed_at":"2026-06-02T01:04:18.226086Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.01008","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:cca8a1bdc3680916a779b9d48926a1806220aa0f4b68c5cbceaa4d985742449d","sha256:89395b16ffa79400b4c934572fcb478122f84feb95cf53e26df79c5059204095"],"state_sha256":"cd0ad1f848a4aa495d9254bf6d35e3dd102751bafe926ed9cf43c7748a34ff47"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mO+g2P6B6EMYl7SHwJH2mqr3ix7V3GrJrHSvCYNI0MvPw7zQfh3t2pGBClv0RGZ08KYUzLTQX568Pff1oDZOCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-29T18:26:00.342948Z","bundle_sha256":"85bc2b4f13ac175a83745982920d1ffd6139b8fea9260f35e88991698ca7f6ef"}}