{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:3QNKTWBJCTWHPN6QR72YPL2LW7","short_pith_number":"pith:3QNKTWBJ","schema_version":"1.0","canonical_sha256":"dc1aa9d82914ec77b7d08ff587af4bb7e90f203dcf30a7f82596fd94ff85b12e","source":{"kind":"arxiv","id":"2504.15476","version":1},"attestation_state":"computed","paper":{"title":"From Reviews to Dialogues: Active Synthesis for Zero-Shot LLM-based Conversational Recommender System","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Dawen Liang, Harald Steck, Julian McAuley, Junda Wu, Nathan Kallus, Rohan Surana, Yu Xia, Zhouhang Xie","submitted_at":"2025-04-21T23:05:47Z","abstract_excerpt":"Conversational recommender systems (CRS) typically require extensive domain-specific conversational datasets, yet high costs, privacy concerns, and data-collection challenges severely limit their availability. Although Large Language Models (LLMs) demonstrate strong zero-shot recommendation capabilities, practical applications often favor smaller, internally managed recommender models due to scalability, interpretability, and data privacy constraints, especially in sensitive or rapidly evolving domains. However, training these smaller models effectively still demands substantial domain-specifi"},"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":"2504.15476","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2025-04-21T23:05:47Z","cross_cats_sorted":[],"title_canon_sha256":"e559f40db25f5e720c55f44244d4d1e5ef5de08650d10b1a4af5fe92279022ac","abstract_canon_sha256":"a2491013afb51c9547532f3b480e829d53c9b025b660fedf0d8e59bf7c977c76"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:52:25.064444Z","signature_b64":"jd7TK+EdqagB2zmtNFtAAKIf6c/eNniSmD5x9QY3FjAjkq/oJ0cAjwPqlqsEGJ/cv2gLgkaTzJW8sUsdnPLDBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dc1aa9d82914ec77b7d08ff587af4bb7e90f203dcf30a7f82596fd94ff85b12e","last_reissued_at":"2026-07-05T10:52:25.063946Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:52:25.063946Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"From Reviews to Dialogues: Active Synthesis for Zero-Shot LLM-based Conversational Recommender System","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Dawen Liang, Harald Steck, Julian McAuley, Junda Wu, Nathan Kallus, Rohan Surana, Yu Xia, Zhouhang Xie","submitted_at":"2025-04-21T23:05:47Z","abstract_excerpt":"Conversational recommender systems (CRS) typically require extensive domain-specific conversational datasets, yet high costs, privacy concerns, and data-collection challenges severely limit their availability. Although Large Language Models (LLMs) demonstrate strong zero-shot recommendation capabilities, practical applications often favor smaller, internally managed recommender models due to scalability, interpretability, and data privacy constraints, especially in sensitive or rapidly evolving domains. However, training these smaller models effectively still demands substantial domain-specifi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.15476","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/2504.15476/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":"2504.15476","created_at":"2026-07-05T10:52:25.064003+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.15476v1","created_at":"2026-07-05T10:52:25.064003+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.15476","created_at":"2026-07-05T10:52:25.064003+00:00"},{"alias_kind":"pith_short_12","alias_value":"3QNKTWBJCTWH","created_at":"2026-07-05T10:52:25.064003+00:00"},{"alias_kind":"pith_short_16","alias_value":"3QNKTWBJCTWHPN6Q","created_at":"2026-07-05T10:52:25.064003+00:00"},{"alias_kind":"pith_short_8","alias_value":"3QNKTWBJ","created_at":"2026-07-05T10:52:25.064003+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.12995","citing_title":"F-GRPO: Factorized Group-Relative Policy Optimization for Unified Candidate Generation and Ranking","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11169","citing_title":"OLIVIA: Online Learning via Inference-time Action Adaptation for Decision Making in LLM ReAct Agents","ref_index":57,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08526","citing_title":"Skill-CMIB: Multimodal Agent Skill for Consistent Action via Conditional Multimodal Information Bottleneck","ref_index":55,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04457","citing_title":"Retrieval Augmented Conversational Recommendation with Reinforcement Learning","ref_index":54,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3QNKTWBJCTWHPN6QR72YPL2LW7","json":"https://pith.science/pith/3QNKTWBJCTWHPN6QR72YPL2LW7.json","graph_json":"https://pith.science/api/pith-number/3QNKTWBJCTWHPN6QR72YPL2LW7/graph.json","events_json":"https://pith.science/api/pith-number/3QNKTWBJCTWHPN6QR72YPL2LW7/events.json","paper":"https://pith.science/paper/3QNKTWBJ"},"agent_actions":{"view_html":"https://pith.science/pith/3QNKTWBJCTWHPN6QR72YPL2LW7","download_json":"https://pith.science/pith/3QNKTWBJCTWHPN6QR72YPL2LW7.json","view_paper":"https://pith.science/paper/3QNKTWBJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.15476&json=true","fetch_graph":"https://pith.science/api/pith-number/3QNKTWBJCTWHPN6QR72YPL2LW7/graph.json","fetch_events":"https://pith.science/api/pith-number/3QNKTWBJCTWHPN6QR72YPL2LW7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3QNKTWBJCTWHPN6QR72YPL2LW7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3QNKTWBJCTWHPN6QR72YPL2LW7/action/storage_attestation","attest_author":"https://pith.science/pith/3QNKTWBJCTWHPN6QR72YPL2LW7/action/author_attestation","sign_citation":"https://pith.science/pith/3QNKTWBJCTWHPN6QR72YPL2LW7/action/citation_signature","submit_replication":"https://pith.science/pith/3QNKTWBJCTWHPN6QR72YPL2LW7/action/replication_record"}},"created_at":"2026-07-05T10:52:25.064003+00:00","updated_at":"2026-07-05T10:52:25.064003+00:00"}