{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:YTNDGNAZLCNYRKC7YPQ2Q2YF3T","short_pith_number":"pith:YTNDGNAZ","schema_version":"1.0","canonical_sha256":"c4da333419589b88a85fc3e1a86b05dcd413232136d77c73ca942c3eeb2a2d53","source":{"kind":"arxiv","id":"2505.04628","version":1},"attestation_state":"computed","paper":{"title":"How Social is It? A Benchmark for LLMs' Capabilities in Multi-user Multi-turn Social Agent Tasks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.SI"],"primary_cat":"cs.CL","authors_text":"Junwu Xiong, Xiaotie Deng, Yusen Wu","submitted_at":"2025-04-04T08:59:01Z","abstract_excerpt":"Expanding the application of large language models (LLMs) to societal life, instead of primary function only as auxiliary assistants to communicate with only one person at a time, necessitates LLMs' capabilities to independently play roles in multi-user, multi-turn social agent tasks within complex social settings. However, currently the capability has not been systematically measured with available benchmarks. To address this gap, we first introduce an agent task leveling framework grounded in sociological principles. Concurrently, we propose a novel benchmark, How Social Is It (we call it HS"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2505.04628","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-04T08:59:01Z","cross_cats_sorted":["cs.AI","cs.SI"],"title_canon_sha256":"435eef4624c3f3892c33ce138084c36faf7041abe0f05f422a71ebd4a561335c","abstract_canon_sha256":"6e7a116e409fc40ade54b53f276d1d6703a41a0d90106d172ca0f06ba4a502ea"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:00:10.491807Z","signature_b64":"AQmR4EVbv8Xr//n356l68vVP0SYct0yqI+7LP5Hg8FiYFHuIHmJoW75kq6yh+KGA5PyckzZtlzFHh2eEo0c+Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c4da333419589b88a85fc3e1a86b05dcd413232136d77c73ca942c3eeb2a2d53","last_reissued_at":"2026-07-05T11:00:10.491292Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:00:10.491292Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"How Social is It? A Benchmark for LLMs' Capabilities in Multi-user Multi-turn Social Agent Tasks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.SI"],"primary_cat":"cs.CL","authors_text":"Junwu Xiong, Xiaotie Deng, Yusen Wu","submitted_at":"2025-04-04T08:59:01Z","abstract_excerpt":"Expanding the application of large language models (LLMs) to societal life, instead of primary function only as auxiliary assistants to communicate with only one person at a time, necessitates LLMs' capabilities to independently play roles in multi-user, multi-turn social agent tasks within complex social settings. However, currently the capability has not been systematically measured with available benchmarks. To address this gap, we first introduce an agent task leveling framework grounded in sociological principles. Concurrently, we propose a novel benchmark, How Social Is It (we call it HS"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.04628","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2505.04628/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2505.04628","created_at":"2026-07-05T11:00:10.491367+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.04628v1","created_at":"2026-07-05T11:00:10.491367+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.04628","created_at":"2026-07-05T11:00:10.491367+00:00"},{"alias_kind":"pith_short_12","alias_value":"YTNDGNAZLCNY","created_at":"2026-07-05T11:00:10.491367+00:00"},{"alias_kind":"pith_short_16","alias_value":"YTNDGNAZLCNYRKC7","created_at":"2026-07-05T11:00:10.491367+00:00"},{"alias_kind":"pith_short_8","alias_value":"YTNDGNAZ","created_at":"2026-07-05T11:00:10.491367+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YTNDGNAZLCNYRKC7YPQ2Q2YF3T","json":"https://pith.science/pith/YTNDGNAZLCNYRKC7YPQ2Q2YF3T.json","graph_json":"https://pith.science/api/pith-number/YTNDGNAZLCNYRKC7YPQ2Q2YF3T/graph.json","events_json":"https://pith.science/api/pith-number/YTNDGNAZLCNYRKC7YPQ2Q2YF3T/events.json","paper":"https://pith.science/paper/YTNDGNAZ"},"agent_actions":{"view_html":"https://pith.science/pith/YTNDGNAZLCNYRKC7YPQ2Q2YF3T","download_json":"https://pith.science/pith/YTNDGNAZLCNYRKC7YPQ2Q2YF3T.json","view_paper":"https://pith.science/paper/YTNDGNAZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.04628&json=true","fetch_graph":"https://pith.science/api/pith-number/YTNDGNAZLCNYRKC7YPQ2Q2YF3T/graph.json","fetch_events":"https://pith.science/api/pith-number/YTNDGNAZLCNYRKC7YPQ2Q2YF3T/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YTNDGNAZLCNYRKC7YPQ2Q2YF3T/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YTNDGNAZLCNYRKC7YPQ2Q2YF3T/action/storage_attestation","attest_author":"https://pith.science/pith/YTNDGNAZLCNYRKC7YPQ2Q2YF3T/action/author_attestation","sign_citation":"https://pith.science/pith/YTNDGNAZLCNYRKC7YPQ2Q2YF3T/action/citation_signature","submit_replication":"https://pith.science/pith/YTNDGNAZLCNYRKC7YPQ2Q2YF3T/action/replication_record"}},"created_at":"2026-07-05T11:00:10.491367+00:00","updated_at":"2026-07-05T11:00:10.491367+00:00"}