{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:SCX6FV4EBSZ6D2RJAIBQMXK2HD","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":"ff61349877992d34b3bed4dd593aa89c276454cfaca39290c88087754e2483d9","cross_cats_sorted":["cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-08T17:55:04Z","title_canon_sha256":"90e93752392846366c8cb03bb7c4b43f289b77939282d77b71084038d2de81fd"},"schema_version":"1.0","source":{"id":"2407.06249","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.06249","created_at":"2026-07-05T10:43:36Z"},{"alias_kind":"arxiv_version","alias_value":"2407.06249v3","created_at":"2026-07-05T10:43:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.06249","created_at":"2026-07-05T10:43:36Z"},{"alias_kind":"pith_short_12","alias_value":"SCX6FV4EBSZ6","created_at":"2026-07-05T10:43:36Z"},{"alias_kind":"pith_short_16","alias_value":"SCX6FV4EBSZ6D2RJ","created_at":"2026-07-05T10:43:36Z"},{"alias_kind":"pith_short_8","alias_value":"SCX6FV4E","created_at":"2026-07-05T10:43:36Z"}],"graph_snapshots":[{"event_id":"sha256:5b9ad8054dafaf637e74cd85f3ed88d02f2d3eeb73aed99a016da2bb7ed82e70","target":"graph","created_at":"2026-07-05T10:43:36Z","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/2407.06249/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) are increasingly being used to synthesize and reason about source code. However, the static nature of these models' knowledge does not reflect the fact that libraries and API functions they invoke are continuously evolving, with functionality being added or changing. While numerous benchmarks evaluate how LLMs can generate code, no prior work has studied how an LLMs' knowledge about code API functions can be updated. To fill this gap, we present CodeUpdateArena, a benchmark for knowledge editing in the code domain. An instance in our benchmark consists of a synthet","authors_text":"Eunsol Choi, Greg Durrett, Shrey Pandit, Xi Ye, Zeyu Leo Liu","cross_cats":["cs.SE"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-08T17:55:04Z","title":"CodeUpdateArena: Benchmarking Knowledge Editing on API Updates"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.06249","kind":"arxiv","version":3},"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:aad426cbc2e514d9013b3e6ffe847dcd57f7ceb1c19b9e5bf09dde626a559f79","target":"record","created_at":"2026-07-05T10:43:36Z","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":"ff61349877992d34b3bed4dd593aa89c276454cfaca39290c88087754e2483d9","cross_cats_sorted":["cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-08T17:55:04Z","title_canon_sha256":"90e93752392846366c8cb03bb7c4b43f289b77939282d77b71084038d2de81fd"},"schema_version":"1.0","source":{"id":"2407.06249","kind":"arxiv","version":3}},"canonical_sha256":"90afe2d7840cb3e1ea290203065d5a38f40b15ea1056571e998cd10d518d4a92","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"90afe2d7840cb3e1ea290203065d5a38f40b15ea1056571e998cd10d518d4a92","first_computed_at":"2026-07-05T10:43:36.629610Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:43:36.629610Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"vBsDq7WpYiiC9HePaBPV0DWl3KEMDCQVzzjlMXiMb3L0nmBqVtOkpqKf3i13vIlnq6wXi9iLqvRiWO9X32XVBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:43:36.630064Z","signed_message":"canonical_sha256_bytes"},"source_id":"2407.06249","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:aad426cbc2e514d9013b3e6ffe847dcd57f7ceb1c19b9e5bf09dde626a559f79","sha256:5b9ad8054dafaf637e74cd85f3ed88d02f2d3eeb73aed99a016da2bb7ed82e70"],"state_sha256":"f8d0ed678ee74a96e3e23b7c57973222d3e3454beaa3eb455bbe8930522adda2"}