{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:ZNJ4BBU62XRONBZO6AYZY5MZRW","short_pith_number":"pith:ZNJ4BBU6","canonical_record":{"source":{"id":"2607.10277","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-07-11T12:17:49Z","cross_cats_sorted":[],"title_canon_sha256":"3b74725dd95f1cfc707ba12b2065d6ff64b35a51c76eb1572c5f463704857b97","abstract_canon_sha256":"7e1b0b6745988c341a5f20bb3543968ff731fd0d89983aae6f4fa9f5d550a223"},"schema_version":"1.0"},"canonical_sha256":"cb53c0869ed5e2e6872ef0319c75998d9a5eef9b0f35874249c147d98ab726ba","source":{"kind":"arxiv","id":"2607.10277","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.10277","created_at":"2026-07-14T01:20:34Z"},{"alias_kind":"arxiv_version","alias_value":"2607.10277v1","created_at":"2026-07-14T01:20:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.10277","created_at":"2026-07-14T01:20:34Z"},{"alias_kind":"pith_short_12","alias_value":"ZNJ4BBU62XRO","created_at":"2026-07-14T01:20:34Z"},{"alias_kind":"pith_short_16","alias_value":"ZNJ4BBU62XRONBZO","created_at":"2026-07-14T01:20:34Z"},{"alias_kind":"pith_short_8","alias_value":"ZNJ4BBU6","created_at":"2026-07-14T01:20:34Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:ZNJ4BBU62XRONBZO6AYZY5MZRW","target":"record","payload":{"canonical_record":{"source":{"id":"2607.10277","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-07-11T12:17:49Z","cross_cats_sorted":[],"title_canon_sha256":"3b74725dd95f1cfc707ba12b2065d6ff64b35a51c76eb1572c5f463704857b97","abstract_canon_sha256":"7e1b0b6745988c341a5f20bb3543968ff731fd0d89983aae6f4fa9f5d550a223"},"schema_version":"1.0"},"canonical_sha256":"cb53c0869ed5e2e6872ef0319c75998d9a5eef9b0f35874249c147d98ab726ba","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-14T01:20:34.223394Z","signature_b64":"4Q3fc4C2wMU/+WQGxcevapKg/gghuL244Sy3dYlPkLk91pkNtxm3l4da1Ox639Gx5/LkUShI+71G3hZcO8wGDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cb53c0869ed5e2e6872ef0319c75998d9a5eef9b0f35874249c147d98ab726ba","last_reissued_at":"2026-07-14T01:20:34.222516Z","signature_status":"signed_v1","first_computed_at":"2026-07-14T01:20:34.222516Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.10277","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":"2026-07-14T01:20:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"u4YMjwG92ayraEPHzncjAyHZmKqd5ACRUjb8MBsml78DcElHoxBrNRImt3LEqyIL3kaP/eku5HfuLRBehy3cDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T08:09:00.131879Z"},"content_sha256":"3c10eac21abb23bb84bdbecfd0fa13ed7ec1ace35b92d094ff1f0f7cea11e13b","schema_version":"1.0","event_id":"sha256:3c10eac21abb23bb84bdbecfd0fa13ed7ec1ace35b92d094ff1f0f7cea11e13b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:ZNJ4BBU62XRONBZO6AYZY5MZRW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"From Business Requirements to Test Assertions: Evaluating LLM-Generated Oracles on Real Bugs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.SE","authors_text":"Nasir U. Eisty, Tiancheng Ma","submitted_at":"2026-07-11T12:17:49Z","abstract_excerpt":"The oracle problem (determining the correct expected outcome for a test) remains a major bottleneck in automated testing, and is increasingly relevant as non-experts rely on AI-generated code they cannot reliably validate. We study whether large language models (LLMs) can generate generalizable test oracles directly from natural-language business requirements, without access to source code or example input-output pairs. We propose a reproducible, requirement-driven pipeline grounded in Defects4J. For each of 10 real bugs from Defects4J Lang (Bugs 1 and 3-11), we (i) extract behavioral changes "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.10277","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/2607.10277/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":"2026-07-14T01:20:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"k3OEOLybpHMUOJDx1Jus4s9S/nIyjUBlEBNhVGdtmaOu45X8FXk3tDqeHIoQwX7YrlpDTDLR5pwfH0TLno5cAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T08:09:00.132392Z"},"content_sha256":"afa3ec46ccc7e3cae7c0533eaef4a9bad24115f04a0540fbdc2f45597d38cf5d","schema_version":"1.0","event_id":"sha256:afa3ec46ccc7e3cae7c0533eaef4a9bad24115f04a0540fbdc2f45597d38cf5d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZNJ4BBU62XRONBZO6AYZY5MZRW/bundle.json","state_url":"https://pith.science/pith/ZNJ4BBU62XRONBZO6AYZY5MZRW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZNJ4BBU62XRONBZO6AYZY5MZRW/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":"2026-08-07T08:09:00Z","links":{"resolver":"https://pith.science/pith/ZNJ4BBU62XRONBZO6AYZY5MZRW","bundle":"https://pith.science/pith/ZNJ4BBU62XRONBZO6AYZY5MZRW/bundle.json","state":"https://pith.science/pith/ZNJ4BBU62XRONBZO6AYZY5MZRW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZNJ4BBU62XRONBZO6AYZY5MZRW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:ZNJ4BBU62XRONBZO6AYZY5MZRW","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":"7e1b0b6745988c341a5f20bb3543968ff731fd0d89983aae6f4fa9f5d550a223","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-07-11T12:17:49Z","title_canon_sha256":"3b74725dd95f1cfc707ba12b2065d6ff64b35a51c76eb1572c5f463704857b97"},"schema_version":"1.0","source":{"id":"2607.10277","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.10277","created_at":"2026-07-14T01:20:34Z"},{"alias_kind":"arxiv_version","alias_value":"2607.10277v1","created_at":"2026-07-14T01:20:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.10277","created_at":"2026-07-14T01:20:34Z"},{"alias_kind":"pith_short_12","alias_value":"ZNJ4BBU62XRO","created_at":"2026-07-14T01:20:34Z"},{"alias_kind":"pith_short_16","alias_value":"ZNJ4BBU62XRONBZO","created_at":"2026-07-14T01:20:34Z"},{"alias_kind":"pith_short_8","alias_value":"ZNJ4BBU6","created_at":"2026-07-14T01:20:34Z"}],"graph_snapshots":[{"event_id":"sha256:afa3ec46ccc7e3cae7c0533eaef4a9bad24115f04a0540fbdc2f45597d38cf5d","target":"graph","created_at":"2026-07-14T01:20:34Z","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/2607.10277/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The oracle problem (determining the correct expected outcome for a test) remains a major bottleneck in automated testing, and is increasingly relevant as non-experts rely on AI-generated code they cannot reliably validate. We study whether large language models (LLMs) can generate generalizable test oracles directly from natural-language business requirements, without access to source code or example input-output pairs. We propose a reproducible, requirement-driven pipeline grounded in Defects4J. For each of 10 real bugs from Defects4J Lang (Bugs 1 and 3-11), we (i) extract behavioral changes ","authors_text":"Nasir U. Eisty, Tiancheng Ma","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-07-11T12:17:49Z","title":"From Business Requirements to Test Assertions: Evaluating LLM-Generated Oracles on Real Bugs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.10277","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":"sha256:3c10eac21abb23bb84bdbecfd0fa13ed7ec1ace35b92d094ff1f0f7cea11e13b","target":"record","created_at":"2026-07-14T01:20:34Z","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":"7e1b0b6745988c341a5f20bb3543968ff731fd0d89983aae6f4fa9f5d550a223","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-07-11T12:17:49Z","title_canon_sha256":"3b74725dd95f1cfc707ba12b2065d6ff64b35a51c76eb1572c5f463704857b97"},"schema_version":"1.0","source":{"id":"2607.10277","kind":"arxiv","version":1}},"canonical_sha256":"cb53c0869ed5e2e6872ef0319c75998d9a5eef9b0f35874249c147d98ab726ba","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cb53c0869ed5e2e6872ef0319c75998d9a5eef9b0f35874249c147d98ab726ba","first_computed_at":"2026-07-14T01:20:34.222516Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-14T01:20:34.222516Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"4Q3fc4C2wMU/+WQGxcevapKg/gghuL244Sy3dYlPkLk91pkNtxm3l4da1Ox639Gx5/LkUShI+71G3hZcO8wGDA==","signature_status":"signed_v1","signed_at":"2026-07-14T01:20:34.223394Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.10277","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3c10eac21abb23bb84bdbecfd0fa13ed7ec1ace35b92d094ff1f0f7cea11e13b","sha256:afa3ec46ccc7e3cae7c0533eaef4a9bad24115f04a0540fbdc2f45597d38cf5d"],"state_sha256":"aa88a76a0d6d62ef8c81c003f2324ef9254b4a0530f02352dc7c8e1425689131"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lLRD1q7ebHU3V1CbxZ2ycbCUExW1Yi++gXzJUWWGbdqa+VKUjyyFysXr1/kQAhKNYEDvntSEmR5XmPQ7NfFNDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T08:09:00.137834Z","bundle_sha256":"af4823c1e0facc24b85128d276f5f8ec1c8188f696d89d04d6dd2e063d5e7357"}}