{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:SMBNTNYC2P6LKNHSYWF6GHYMFF","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":"6d8cd3e395ab9dedfa84f80c95a5cdaf65ae9a9a01b4c64dced1a2ebe3df9059","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-24T09:21:06Z","title_canon_sha256":"0d8327e6d0c2862a9aaa25ddd0b18b6a769cacf2ed95fb4ec95e642e3e49354d"},"schema_version":"1.0","source":{"id":"2305.14938","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.14938","created_at":"2026-07-05T08:02:43Z"},{"alias_kind":"arxiv_version","alias_value":"2305.14938v2","created_at":"2026-07-05T08:02:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.14938","created_at":"2026-07-05T08:02:43Z"},{"alias_kind":"pith_short_12","alias_value":"SMBNTNYC2P6L","created_at":"2026-07-05T08:02:43Z"},{"alias_kind":"pith_short_16","alias_value":"SMBNTNYC2P6LKNHS","created_at":"2026-07-05T08:02:43Z"},{"alias_kind":"pith_short_8","alias_value":"SMBNTNYC","created_at":"2026-07-05T08:02:43Z"}],"graph_snapshots":[{"event_id":"sha256:7b7acec903139df38a7bbe717d782249d74671311017bec57a9586467163231f","target":"graph","created_at":"2026-07-05T08:02:43Z","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/2305.14938/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) have been shown to perform well at a variety of syntactic, discourse, and reasoning tasks. While LLMs are increasingly deployed in many forms including conversational agents that interact with humans, we lack a grounded benchmark to measure how well LLMs understand \\textit{social} language. Here, we introduce a new theory-driven benchmark, SocKET, that contains 58 NLP tasks testing social knowledge which we group into five categories: humor & sarcasm, offensiveness, sentiment & emotion, and trustworthiness. In tests on the benchmark, we demonstrate that current mod","authors_text":"Chang Shu, David Jurgens, Jiaxin Pei, Minje Choi, Sagar Kumar","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-24T09:21:06Z","title":"Do LLMs Understand Social Knowledge? Evaluating the Sociability of Large Language Models with SocKET Benchmark"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.14938","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":"sha256:99cb924741479f07840540cfe36bedf3b1ff9e64f1740b99787205f99ae5ed53","target":"record","created_at":"2026-07-05T08:02:43Z","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":"6d8cd3e395ab9dedfa84f80c95a5cdaf65ae9a9a01b4c64dced1a2ebe3df9059","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-24T09:21:06Z","title_canon_sha256":"0d8327e6d0c2862a9aaa25ddd0b18b6a769cacf2ed95fb4ec95e642e3e49354d"},"schema_version":"1.0","source":{"id":"2305.14938","kind":"arxiv","version":2}},"canonical_sha256":"9302d9b702d3fcb534f2c58be31f0c294f86c097413448a91db69e7ed8437c58","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9302d9b702d3fcb534f2c58be31f0c294f86c097413448a91db69e7ed8437c58","first_computed_at":"2026-07-05T08:02:43.761195Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:02:43.761195Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"6DT0kMKwmGioz084e/ohbubDZtZ7WexZdFv/LHzIheUzLdLiDj5OImzfbTITf5o/DHeJIPWtGjVkNrN1cbLhAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:02:43.761706Z","signed_message":"canonical_sha256_bytes"},"source_id":"2305.14938","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:99cb924741479f07840540cfe36bedf3b1ff9e64f1740b99787205f99ae5ed53","sha256:7b7acec903139df38a7bbe717d782249d74671311017bec57a9586467163231f"],"state_sha256":"da0fa315a2705b6832f1765b58a2bcd1b46bfae1b9c6acca6c1f6d0ce7828ede"}