{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:TDL7NEHGHB6LWH4K6BL7RWHCSI","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":"7ca77badffa73000dcb76f0b1b95c5902272f635126e75ce040cc76bcdf2fdad","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2026-04-02T01:13:40Z","title_canon_sha256":"5b734335a9b3a34e7b65c423fb5c4cde7eb173b86c0571d8a99e8a6b47835261"},"schema_version":"1.0","source":{"id":"2604.01518","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.01518","created_at":"2026-08-07T00:45:48Z"},{"alias_kind":"arxiv_version","alias_value":"2604.01518v2","created_at":"2026-08-07T00:45:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.01518","created_at":"2026-08-07T00:45:48Z"},{"alias_kind":"pith_short_12","alias_value":"TDL7NEHGHB6L","created_at":"2026-08-07T00:45:48Z"},{"alias_kind":"pith_short_16","alias_value":"TDL7NEHGHB6LWH4K","created_at":"2026-08-07T00:45:48Z"},{"alias_kind":"pith_short_8","alias_value":"TDL7NEHG","created_at":"2026-08-07T00:45:48Z"}],"graph_snapshots":[{"event_id":"sha256:00acb3ac9680e9d24372ea4500346a643262d91f1c4f0690824d0d2e897f0b3e","target":"graph","created_at":"2026-08-07T00:45:48Z","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/2604.01518/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Test-based benchmarks such as SWE-bench have become a standard basis for evaluating automated issue resolution agents, deeming a patch correct if it passes a provided regression test suite. In practice, weak test suites can admit plausible but semantically incorrect patches, inflating reported agent performance. We present \\tool, a test augmentation framework that uses semantically modified program variants as behavioral probes to identify and close gaps in benchmark test suites. Variants of the reference patch that survive the original tests reveal under-constrained behaviors, which then guid","authors_text":"Chenglin Li, Shin Hwei Tan, Tse-Hsun (Peter) Chen, Yisen Xu, Zehao Wang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2026-04-02T01:13:40Z","title":"Probe to Generate: Program Variant-Guided Test Augmentation for Repository-Level Repair Benchmarks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.01518","kind":"arxiv","version":2},"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:df28942164a833996d1a58837b609951932c9996badb5189c250f2c23fe22a66","target":"record","created_at":"2026-08-07T00:45:48Z","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":"7ca77badffa73000dcb76f0b1b95c5902272f635126e75ce040cc76bcdf2fdad","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2026-04-02T01:13:40Z","title_canon_sha256":"5b734335a9b3a34e7b65c423fb5c4cde7eb173b86c0571d8a99e8a6b47835261"},"schema_version":"1.0","source":{"id":"2604.01518","kind":"arxiv","version":2}},"canonical_sha256":"98d7f690e6387cbb1f8af057f8d8e2920cf7c2498b53b188f21dda690f475297","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"98d7f690e6387cbb1f8af057f8d8e2920cf7c2498b53b188f21dda690f475297","first_computed_at":"2026-08-07T00:45:48.726897Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-07T00:45:48.726897Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"sAJ1apkzo9qVhBP389vqqPidvo5WRC2vxCSAGN9adAJyUDPdasK5sctXPjbvObl+w/IhTmu76DwnxmWz+5d3BA==","signature_status":"signed_v1","signed_at":"2026-08-07T00:45:48.728331Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.01518","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:df28942164a833996d1a58837b609951932c9996badb5189c250f2c23fe22a66","sha256:00acb3ac9680e9d24372ea4500346a643262d91f1c4f0690824d0d2e897f0b3e"],"state_sha256":"48544a90e51e0b132cf7989314226c0fdd268d3afd7ed317f6e5ff3f203656f0"}