{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:6KAKEGFQRBV7KPXVYVGBFEQTAC","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":"e6733b2b9d386c80cb456416da0684b648542140a45850000043492e223cf43e","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2025-03-06T09:22:23Z","title_canon_sha256":"0f8a5aaab2fa549e156ca080fa7775e3dc121f3dccde36f916bc3d4180748f8b"},"schema_version":"1.0","source":{"id":"2503.04241","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.04241","created_at":"2026-07-05T10:25:33Z"},{"alias_kind":"arxiv_version","alias_value":"2503.04241v1","created_at":"2026-07-05T10:25:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.04241","created_at":"2026-07-05T10:25:33Z"},{"alias_kind":"pith_short_12","alias_value":"6KAKEGFQRBV7","created_at":"2026-07-05T10:25:33Z"},{"alias_kind":"pith_short_16","alias_value":"6KAKEGFQRBV7KPXV","created_at":"2026-07-05T10:25:33Z"},{"alias_kind":"pith_short_8","alias_value":"6KAKEGFQ","created_at":"2026-07-05T10:25:33Z"}],"graph_snapshots":[{"event_id":"sha256:0e841382afed22cfa87007e2b5b16623def1168df109448ab08d54696d7785ac","target":"graph","created_at":"2026-07-05T10:25:33Z","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/2503.04241/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Modern Large Language Models (LLMs) have shown astounding capabilities of code understanding and synthesis. In order to assess such capabilities, several benchmarks have been devised (e.g., HumanEval). However, most benchmarks focus on code synthesis from natural language instructions. Hence, such benchmarks do not test for other forms of code understanding. Moreover, there have been concerns about contamination and leakage. That is, benchmark problems (or closely related problems) may appear in training set, strongly biasing benchmark results. In this work we investigate whether large languag","authors_text":"Julian Aron Prenner, Romain Robbes","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2025-03-06T09:22:23Z","title":"ThrowBench: Benchmarking LLMs by Predicting Runtime Exceptions"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.04241","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":"sha256:85e8c747d6174aa5a631906e31dba47ddb64f1c9ba84074fc4b83c0092ed5b4d","target":"record","created_at":"2026-07-05T10:25:33Z","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":"e6733b2b9d386c80cb456416da0684b648542140a45850000043492e223cf43e","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2025-03-06T09:22:23Z","title_canon_sha256":"0f8a5aaab2fa549e156ca080fa7775e3dc121f3dccde36f916bc3d4180748f8b"},"schema_version":"1.0","source":{"id":"2503.04241","kind":"arxiv","version":1}},"canonical_sha256":"f280a218b0886bf53ef5c54c129213008bb2e8dc793fcbfd57462ea8c2d856a7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f280a218b0886bf53ef5c54c129213008bb2e8dc793fcbfd57462ea8c2d856a7","first_computed_at":"2026-07-05T10:25:33.239054Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:25:33.239054Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Dywo23kdej5UGE8vJ3YuaBHwUZZnHvcEXDx3fuYDShiT4LGi/vCTyu0mzOn/2QX+ALDaYwaz8/MmCmljz6LvBA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:25:33.239585Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.04241","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:85e8c747d6174aa5a631906e31dba47ddb64f1c9ba84074fc4b83c0092ed5b4d","sha256:0e841382afed22cfa87007e2b5b16623def1168df109448ab08d54696d7785ac"],"state_sha256":"946a41ad08b65f54720b3a9c7a46838a3b79854405c46216205579c2c039756e"}