{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:RC2WKF3UDUMDJLT3GKBZJK3KDB","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":"da6c93ad7f3aaaaeb7c0172a9ff7fff05a811ef025ee98d468b179112b09d0ab","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-09-27T13:02:06Z","title_canon_sha256":"c383810179fe54a3bb6f8a0a4751bcf68b118ae9416a75a713a23bcdc95493eb"},"schema_version":"1.0","source":{"id":"2309.15630","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2309.15630","created_at":"2026-07-05T07:02:32Z"},{"alias_kind":"arxiv_version","alias_value":"2309.15630v4","created_at":"2026-07-05T07:02:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.15630","created_at":"2026-07-05T07:02:32Z"},{"alias_kind":"pith_short_12","alias_value":"RC2WKF3UDUMD","created_at":"2026-07-05T07:02:32Z"},{"alias_kind":"pith_short_16","alias_value":"RC2WKF3UDUMDJLT3","created_at":"2026-07-05T07:02:32Z"},{"alias_kind":"pith_short_8","alias_value":"RC2WKF3U","created_at":"2026-07-05T07:02:32Z"}],"graph_snapshots":[{"event_id":"sha256:d6961389b08e1338e48b90a307d8af7f84190e4f79ef587e523bb45c67ca58b3","target":"graph","created_at":"2026-07-05T07:02:32Z","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/2309.15630/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent developments in large language models (LLMs) have shown promise in enhancing the capabilities of natural language processing (NLP). Despite these successes, there remains a dearth of research dedicated to the NLP problem-solving abilities of LLMs. To fill the gap in this area, we present a unique benchmarking dataset, NLPBench, comprising 378 college-level NLP questions spanning various NLP topics sourced from Yale University's prior final exams. NLPBench includes questions with context, in which multiple sub-questions share the same public information, and diverse question types, inclu","authors_text":"Irene Li, Jieyu Zhang, Lechao Cheng, Linxin Song, Pengyuan Zhou, Tianyi Zhou","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-09-27T13:02:06Z","title":"NLPBench: Evaluating Large Language Models on Solving NLP Problems"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.15630","kind":"arxiv","version":4},"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:f7632b2a2dccb77d05e549a7713034b67807a04735e0298fdbacbc766d1402df","target":"record","created_at":"2026-07-05T07:02:32Z","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":"da6c93ad7f3aaaaeb7c0172a9ff7fff05a811ef025ee98d468b179112b09d0ab","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-09-27T13:02:06Z","title_canon_sha256":"c383810179fe54a3bb6f8a0a4751bcf68b118ae9416a75a713a23bcdc95493eb"},"schema_version":"1.0","source":{"id":"2309.15630","kind":"arxiv","version":4}},"canonical_sha256":"88b56517741d1834ae7b328394ab6a18570f07109a0fae2490e2024ab98ad0bc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"88b56517741d1834ae7b328394ab6a18570f07109a0fae2490e2024ab98ad0bc","first_computed_at":"2026-07-05T07:02:32.627804Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:02:32.627804Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"p3dNADRkL1UL7weeBDeGkGcGs9W3Z3I9jgPNa/gKdVmPMyrHInKxLUnTP5K7W7ziEoS5TjCNUm1132ebUCEOBg==","signature_status":"signed_v1","signed_at":"2026-07-05T07:02:32.628399Z","signed_message":"canonical_sha256_bytes"},"source_id":"2309.15630","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f7632b2a2dccb77d05e549a7713034b67807a04735e0298fdbacbc766d1402df","sha256:d6961389b08e1338e48b90a307d8af7f84190e4f79ef587e523bb45c67ca58b3"],"state_sha256":"245393cb226a42e647254b6c92dd94e6edc50632319d4acf3e44c9ea78b627c3"}