{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:IVD4CGERAE7HTU6IDS2TMQR75Z","short_pith_number":"pith:IVD4CGER","schema_version":"1.0","canonical_sha256":"4547c11891013e79d3c81cb536423fee4375f1d19785e6b421805dbc92e55895","source":{"kind":"arxiv","id":"2502.00640","version":3},"attestation_state":"computed","paper":{"title":"CollabLLM: From Passive Responders to Active Collaborators","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Baolin Peng, Gavin Li, Hao Cheng, James Zou, Jianfeng Gao, Jure Leskovec, Michel Galley, Shirley Wu, Weixin Cai, Yao Dou","submitted_at":"2025-02-02T03:05:52Z","abstract_excerpt":"Large Language Models are typically trained with next-turn rewards, limiting their ability to optimize for long-term interaction. As a result, they often respond passively to ambiguous or open-ended user requests, failing to help users reach their ultimate intents and leading to inefficient conversations. To address these limitations, we introduce CollabLLM, a novel and general training framework that enhances multiturn human-LLM collaboration. Its key innovation is a collaborative simulation that estimates the long-term contribution of responses using Multiturn-aware Rewards. By reinforcement"},"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":"2502.00640","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-02-02T03:05:52Z","cross_cats_sorted":[],"title_canon_sha256":"2146663ff97eb3b5d89c5256be95a40e99616ff51411aa6a64b36086b1ee83b7","abstract_canon_sha256":"4777d46092fec732826fdfd1495f1a97800b1bd54837e0fb2e64190509249fb5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:45:14.127635Z","signature_b64":"IQISdHUxBx92pr7iIkeWPFgJ7sM3Akrzl4kzfsHBfs0jCsGfUAIneXdsDr0kWESPV3YD7nKpeXp/ls7aLY0YAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4547c11891013e79d3c81cb536423fee4375f1d19785e6b421805dbc92e55895","last_reissued_at":"2026-07-05T11:45:14.127197Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:45:14.127197Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CollabLLM: From Passive Responders to Active Collaborators","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Baolin Peng, Gavin Li, Hao Cheng, James Zou, Jianfeng Gao, Jure Leskovec, Michel Galley, Shirley Wu, Weixin Cai, Yao Dou","submitted_at":"2025-02-02T03:05:52Z","abstract_excerpt":"Large Language Models are typically trained with next-turn rewards, limiting their ability to optimize for long-term interaction. As a result, they often respond passively to ambiguous or open-ended user requests, failing to help users reach their ultimate intents and leading to inefficient conversations. To address these limitations, we introduce CollabLLM, a novel and general training framework that enhances multiturn human-LLM collaboration. Its key innovation is a collaborative simulation that estimates the long-term contribution of responses using Multiturn-aware Rewards. By reinforcement"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.00640","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/2502.00640/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":"2502.00640","created_at":"2026-07-05T11:45:14.127253+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.00640v3","created_at":"2026-07-05T11:45:14.127253+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.00640","created_at":"2026-07-05T11:45:14.127253+00:00"},{"alias_kind":"pith_short_12","alias_value":"IVD4CGERAE7H","created_at":"2026-07-05T11:45:14.127253+00:00"},{"alias_kind":"pith_short_16","alias_value":"IVD4CGERAE7HTU6I","created_at":"2026-07-05T11:45:14.127253+00:00"},{"alias_kind":"pith_short_8","alias_value":"IVD4CGER","created_at":"2026-07-05T11:45:14.127253+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":9,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.06399","citing_title":"CollabSim: A CSCW-Grounded Methodology for Investigating Collaborative Competence of LLM Agents through Controlled Multi-Agent Experiments","ref_index":65,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29495","citing_title":"Cognitive World Model for Progressive BDI/E Trajectory Evaluation of Conversational Agents","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2605.26969","citing_title":"Recon: Reconstruction-Guided Reasoning Synthesis for User Modeling","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16616","citing_title":"MLReplicate: Benchmarking Autonomous Research Systems for Machine Learning Reproducibility","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2601.21972","citing_title":"Learning Decentralized LLM Collaboration with Multi-Agent Actor Critic","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2604.22785","citing_title":"CoFi-PGMA: Counterfactual Policy Gradients under Filtered Feedback for Multi-Agent LLMs","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09808","citing_title":"Quantifying the Utility of User Simulators for Building Collaborative LLM Assistants","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21827","citing_title":"Alignment has a Fantasia Problem","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2604.07645","citing_title":"PRIME: Training Free Proactive Reasoning via Iterative Memory Evolution for User-Centric Agent","ref_index":19,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IVD4CGERAE7HTU6IDS2TMQR75Z","json":"https://pith.science/pith/IVD4CGERAE7HTU6IDS2TMQR75Z.json","graph_json":"https://pith.science/api/pith-number/IVD4CGERAE7HTU6IDS2TMQR75Z/graph.json","events_json":"https://pith.science/api/pith-number/IVD4CGERAE7HTU6IDS2TMQR75Z/events.json","paper":"https://pith.science/paper/IVD4CGER"},"agent_actions":{"view_html":"https://pith.science/pith/IVD4CGERAE7HTU6IDS2TMQR75Z","download_json":"https://pith.science/pith/IVD4CGERAE7HTU6IDS2TMQR75Z.json","view_paper":"https://pith.science/paper/IVD4CGER","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.00640&json=true","fetch_graph":"https://pith.science/api/pith-number/IVD4CGERAE7HTU6IDS2TMQR75Z/graph.json","fetch_events":"https://pith.science/api/pith-number/IVD4CGERAE7HTU6IDS2TMQR75Z/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IVD4CGERAE7HTU6IDS2TMQR75Z/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IVD4CGERAE7HTU6IDS2TMQR75Z/action/storage_attestation","attest_author":"https://pith.science/pith/IVD4CGERAE7HTU6IDS2TMQR75Z/action/author_attestation","sign_citation":"https://pith.science/pith/IVD4CGERAE7HTU6IDS2TMQR75Z/action/citation_signature","submit_replication":"https://pith.science/pith/IVD4CGERAE7HTU6IDS2TMQR75Z/action/replication_record"}},"created_at":"2026-07-05T11:45:14.127253+00:00","updated_at":"2026-07-05T11:45:14.127253+00:00"}