{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:5SSGVMSDHCY2MEWRFS3UT3GH4N","short_pith_number":"pith:5SSGVMSD","schema_version":"1.0","canonical_sha256":"eca46ab24338b1a612d12cb749ecc7e3784a0a194e6e3d7c6225b7ade65db595","source":{"kind":"arxiv","id":"2501.15427","version":2},"attestation_state":"computed","paper":{"title":"OpenCharacter: Training Customizable Role-Playing LLMs with Large-Scale Synthetic Personas","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dian Yu, Dong Yu, Hongming Zhang, Tao Ge, Wenhao Yu, Xiaoyang Wang","submitted_at":"2025-01-26T07:07:01Z","abstract_excerpt":"Customizable role-playing in large language models (LLMs), also known as character generalization, is gaining increasing attention for its versatility and cost-efficiency in developing and deploying role-playing dialogue agents. This study explores a large-scale data synthesis approach to equip LLMs with character generalization capabilities. We begin by synthesizing large-scale character profiles using personas from Persona Hub and then explore two strategies: response rewriting and response generation, to create character-aligned instructional responses. To validate the effectiveness of our "},"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":"2501.15427","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-26T07:07:01Z","cross_cats_sorted":[],"title_canon_sha256":"aa36ed6a2bcacf86373725f328830fc1f41375620dec6e068af5ccd4b0270a9a","abstract_canon_sha256":"b910df16221ec7feec4b103cba403ef5af94cd5fa93193da157e241bbdc6eb2e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:15:48.597636Z","signature_b64":"j6E3WpqSJIYaQolGVVQS/TfpMaqtiI0BXYbwBFzsrZCPO3AxwYa3Pebv4coM6f16+HMRBzWQbqL2bHr9z+SSAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"eca46ab24338b1a612d12cb749ecc7e3784a0a194e6e3d7c6225b7ade65db595","last_reissued_at":"2026-07-05T10:15:48.596999Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:15:48.596999Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"OpenCharacter: Training Customizable Role-Playing LLMs with Large-Scale Synthetic Personas","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dian Yu, Dong Yu, Hongming Zhang, Tao Ge, Wenhao Yu, Xiaoyang Wang","submitted_at":"2025-01-26T07:07:01Z","abstract_excerpt":"Customizable role-playing in large language models (LLMs), also known as character generalization, is gaining increasing attention for its versatility and cost-efficiency in developing and deploying role-playing dialogue agents. This study explores a large-scale data synthesis approach to equip LLMs with character generalization capabilities. We begin by synthesizing large-scale character profiles using personas from Persona Hub and then explore two strategies: response rewriting and response generation, to create character-aligned instructional responses. To validate the effectiveness of our "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.15427","kind":"arxiv","version":2},"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/2501.15427/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":"2501.15427","created_at":"2026-07-05T10:15:48.597081+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.15427v2","created_at":"2026-07-05T10:15:48.597081+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.15427","created_at":"2026-07-05T10:15:48.597081+00:00"},{"alias_kind":"pith_short_12","alias_value":"5SSGVMSDHCY2","created_at":"2026-07-05T10:15:48.597081+00:00"},{"alias_kind":"pith_short_16","alias_value":"5SSGVMSDHCY2MEWR","created_at":"2026-07-05T10:15:48.597081+00:00"},{"alias_kind":"pith_short_8","alias_value":"5SSGVMSD","created_at":"2026-07-05T10:15:48.597081+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.07918","citing_title":"Efficient Safety Alignment of Language Models via Latent Personality Traits","ref_index":48,"is_internal_anchor":true},{"citing_arxiv_id":"2606.25632","citing_title":"Staying In Character: Perspective-Bounded Memory For Book-Based Role-Playing Agents","ref_index":53,"is_internal_anchor":false},{"citing_arxiv_id":"2606.27025","citing_title":"Improving General Role-Playing Agents via Psychology-Grounded Reasoning and Role-Aware Policy Optimization","ref_index":80,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14169","citing_title":"BOOKMARKS: Efficient Active Storyline Memory for Role-playing","ref_index":120,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5SSGVMSDHCY2MEWRFS3UT3GH4N","json":"https://pith.science/pith/5SSGVMSDHCY2MEWRFS3UT3GH4N.json","graph_json":"https://pith.science/api/pith-number/5SSGVMSDHCY2MEWRFS3UT3GH4N/graph.json","events_json":"https://pith.science/api/pith-number/5SSGVMSDHCY2MEWRFS3UT3GH4N/events.json","paper":"https://pith.science/paper/5SSGVMSD"},"agent_actions":{"view_html":"https://pith.science/pith/5SSGVMSDHCY2MEWRFS3UT3GH4N","download_json":"https://pith.science/pith/5SSGVMSDHCY2MEWRFS3UT3GH4N.json","view_paper":"https://pith.science/paper/5SSGVMSD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.15427&json=true","fetch_graph":"https://pith.science/api/pith-number/5SSGVMSDHCY2MEWRFS3UT3GH4N/graph.json","fetch_events":"https://pith.science/api/pith-number/5SSGVMSDHCY2MEWRFS3UT3GH4N/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5SSGVMSDHCY2MEWRFS3UT3GH4N/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5SSGVMSDHCY2MEWRFS3UT3GH4N/action/storage_attestation","attest_author":"https://pith.science/pith/5SSGVMSDHCY2MEWRFS3UT3GH4N/action/author_attestation","sign_citation":"https://pith.science/pith/5SSGVMSDHCY2MEWRFS3UT3GH4N/action/citation_signature","submit_replication":"https://pith.science/pith/5SSGVMSDHCY2MEWRFS3UT3GH4N/action/replication_record"}},"created_at":"2026-07-05T10:15:48.597081+00:00","updated_at":"2026-07-05T10:15:48.597081+00:00"}