{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:KSBRDAI6MTAKYLAAUIIXYHZMEO","short_pith_number":"pith:KSBRDAI6","schema_version":"1.0","canonical_sha256":"548311811e64c0ac2c00a2117c1f2c23b9153886936a113c3601f0f907bc5947","source":{"kind":"arxiv","id":"2310.00746","version":3},"attestation_state":"computed","paper":{"title":"RoleLLM: Benchmarking, Eliciting, and Enhancing Role-Playing Abilities of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Haoran Que, Hongcheng Guo, Jiaheng Liu, Jian Yang, Jie Fu, Junran Peng, Ke Xu, Man Zhang, Ruitong Gan, Stephen W. Huang, Wangchunshu Zhou, Wanli Ouyang, Yuhan Wu, Zehao Ni, Zekun Moore Wang, Zhaoxiang Zhang, Zhongyuan Peng","submitted_at":"2023-10-01T17:52:59Z","abstract_excerpt":"The advent of Large Language Models (LLMs) has paved the way for complex tasks such as role-playing, which enhances user interactions by enabling models to imitate various characters. However, the closed-source nature of state-of-the-art LLMs and their general-purpose training limit role-playing optimization. In this paper, we introduce RoleLLM, a framework to benchmark, elicit, and enhance role-playing abilities in LLMs. RoleLLM comprises four stages: (1) Role Profile Construction for 100 roles; (2) Context-Based Instruction Generation (Context-Instruct) for role-specific knowledge extraction"},"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":"2310.00746","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-01T17:52:59Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"fd882c8c3c64f7361a53ddb8621a913d30ee6584165e65095956529624f06a13","abstract_canon_sha256":"25d0d8ddcc09201f3b738848c972bf803ed82ac224ee70f19d4429d44ad82082"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:33:32.355742Z","signature_b64":"6ZKGPB8V7zIrHKSDbgzlltAMLYhz6fxgQUaNrMlp+kJnj0WzQJt2phGcxVC5Zhls4SLr69CdvxJDm9eINYriAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"548311811e64c0ac2c00a2117c1f2c23b9153886936a113c3601f0f907bc5947","last_reissued_at":"2026-07-05T08:33:32.355182Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:33:32.355182Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RoleLLM: Benchmarking, Eliciting, and Enhancing Role-Playing Abilities of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Haoran Que, Hongcheng Guo, Jiaheng Liu, Jian Yang, Jie Fu, Junran Peng, Ke Xu, Man Zhang, Ruitong Gan, Stephen W. Huang, Wangchunshu Zhou, Wanli Ouyang, Yuhan Wu, Zehao Ni, Zekun Moore Wang, Zhaoxiang Zhang, Zhongyuan Peng","submitted_at":"2023-10-01T17:52:59Z","abstract_excerpt":"The advent of Large Language Models (LLMs) has paved the way for complex tasks such as role-playing, which enhances user interactions by enabling models to imitate various characters. However, the closed-source nature of state-of-the-art LLMs and their general-purpose training limit role-playing optimization. In this paper, we introduce RoleLLM, a framework to benchmark, elicit, and enhance role-playing abilities in LLMs. RoleLLM comprises four stages: (1) Role Profile Construction for 100 roles; (2) Context-Based Instruction Generation (Context-Instruct) for role-specific knowledge extraction"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.00746","kind":"arxiv","version":3},"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/2310.00746/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":"2310.00746","created_at":"2026-07-05T08:33:32.355251+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.00746v3","created_at":"2026-07-05T08:33:32.355251+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.00746","created_at":"2026-07-05T08:33:32.355251+00:00"},{"alias_kind":"pith_short_12","alias_value":"KSBRDAI6MTAK","created_at":"2026-07-05T08:33:32.355251+00:00"},{"alias_kind":"pith_short_16","alias_value":"KSBRDAI6MTAKYLAA","created_at":"2026-07-05T08:33:32.355251+00:00"},{"alias_kind":"pith_short_8","alias_value":"KSBRDAI6","created_at":"2026-07-05T08:33:32.355251+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":13,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.23969","citing_title":"SLAP: Stratified Loss-based Pruning for On-Policy Data-Efficient Instruction Tuning","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24279","citing_title":"ContextEcho: A Benchmark for Persona Drift in Long Agentic-Coding Sessions","ref_index":83,"is_internal_anchor":false},{"citing_arxiv_id":"2605.27914","citing_title":"Does Capability Transfer to Subjective Behavior -- and Would Our Instruments Tell Us? A Self-Evolving, Trust-by-Construction Evaluation Paradigm","ref_index":123,"is_internal_anchor":false},{"citing_arxiv_id":"2305.09620","citing_title":"AI-Augmented Surveys: Leveraging Large Language Models and Surveys for Opinion Prediction","ref_index":100,"is_internal_anchor":false},{"citing_arxiv_id":"2408.09049","citing_title":"Inertia in Moral and Value Judgments of Large Language Models","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2506.04565","citing_title":"From Standalone LLMs to Integrated Intelligence: A Survey of Compound Al Systems","ref_index":191,"is_internal_anchor":false},{"citing_arxiv_id":"2511.08565","citing_title":"Moral Susceptibility and Robustness under Persona Role-Play in Large Language Models","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2601.06352","citing_title":"CARD: Cluster-level Adaptation with Reward-guided Decoding for Personalized Text Generation","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.09557","citing_title":"SPEED-Bench: A Unified and Diverse Benchmark for Speculative Decoding","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2404.13501","citing_title":"A Survey on the Memory Mechanism of Large Language Model based Agents","ref_index":145,"is_internal_anchor":false},{"citing_arxiv_id":"2604.07054","citing_title":"Sell More, Play Less: Benchmarking LLM Realistic Selling Skill","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05939","citing_title":"Context-Value-Action Architecture for Value-Driven Large Language Model Agents","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2604.13804","citing_title":"Character Beyond Speech: Leveraging Role-Playing Evaluation in Audio Large Language Models via Reinforcement Learning","ref_index":34,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KSBRDAI6MTAKYLAAUIIXYHZMEO","json":"https://pith.science/pith/KSBRDAI6MTAKYLAAUIIXYHZMEO.json","graph_json":"https://pith.science/api/pith-number/KSBRDAI6MTAKYLAAUIIXYHZMEO/graph.json","events_json":"https://pith.science/api/pith-number/KSBRDAI6MTAKYLAAUIIXYHZMEO/events.json","paper":"https://pith.science/paper/KSBRDAI6"},"agent_actions":{"view_html":"https://pith.science/pith/KSBRDAI6MTAKYLAAUIIXYHZMEO","download_json":"https://pith.science/pith/KSBRDAI6MTAKYLAAUIIXYHZMEO.json","view_paper":"https://pith.science/paper/KSBRDAI6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.00746&json=true","fetch_graph":"https://pith.science/api/pith-number/KSBRDAI6MTAKYLAAUIIXYHZMEO/graph.json","fetch_events":"https://pith.science/api/pith-number/KSBRDAI6MTAKYLAAUIIXYHZMEO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KSBRDAI6MTAKYLAAUIIXYHZMEO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KSBRDAI6MTAKYLAAUIIXYHZMEO/action/storage_attestation","attest_author":"https://pith.science/pith/KSBRDAI6MTAKYLAAUIIXYHZMEO/action/author_attestation","sign_citation":"https://pith.science/pith/KSBRDAI6MTAKYLAAUIIXYHZMEO/action/citation_signature","submit_replication":"https://pith.science/pith/KSBRDAI6MTAKYLAAUIIXYHZMEO/action/replication_record"}},"created_at":"2026-07-05T08:33:32.355251+00:00","updated_at":"2026-07-05T08:33:32.355251+00:00"}