{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:XWD5XKUG3EUVTFDFFWN4H66XMM","short_pith_number":"pith:XWD5XKUG","schema_version":"1.0","canonical_sha256":"bd87dbaa86d9295994652d9bc3fbd76320cd9a7787bb9f63f806b718ee9e732c","source":{"kind":"arxiv","id":"2502.18439","version":2},"attestation_state":"computed","paper":{"title":"MAPoRL: Multi-Agent Post-Co-Training for Collaborative Large Language Models with Reinforcement Learning","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Asuman Ozdaglar, Chanwoo Park, Joo-Kyung Kim, Kaiqing Zhang, Seungju Han, Xingzhi Guo","submitted_at":"2025-02-25T18:33:48Z","abstract_excerpt":"Leveraging multiple large language models (LLMs) to build collaborative multi-agentic workflows has demonstrated significant potential. However, most previous studies focus on prompting the out-of-the-box LLMs, relying on their innate capability for collaboration, which may not improve LLMs' performance as shown recently. In this paper, we introduce a new post-training paradigm MAPoRL (Multi-Agent Post-co-training for collaborative LLMs with Reinforcement Learning), to explicitly elicit the collaborative behaviors and further unleash the power of multi-agentic LLM frameworks. In MAPoRL, multip"},"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":"2502.18439","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2025-02-25T18:33:48Z","cross_cats_sorted":[],"title_canon_sha256":"34a0365d303a652a7236a17d9152295560b03c249906a134bcc6d3ca812a2ec5","abstract_canon_sha256":"e80aa4e39889b9786cc310ff84fb7cdbc9ead11396f9dd7522b41fa5b056b463"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:35:53.727106Z","signature_b64":"6c40q7HK2zj6ZpGNpKAsQu8MbnrSiImV54b6+sJ8yGXxHT6ZMzcJmIlDGHft+fAxquMDe7WbnxHvHyXdLcf2CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bd87dbaa86d9295994652d9bc3fbd76320cd9a7787bb9f63f806b718ee9e732c","last_reissued_at":"2026-07-05T11:35:53.726627Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:35:53.726627Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MAPoRL: Multi-Agent Post-Co-Training for Collaborative Large Language Models with Reinforcement Learning","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Asuman Ozdaglar, Chanwoo Park, Joo-Kyung Kim, Kaiqing Zhang, Seungju Han, Xingzhi Guo","submitted_at":"2025-02-25T18:33:48Z","abstract_excerpt":"Leveraging multiple large language models (LLMs) to build collaborative multi-agentic workflows has demonstrated significant potential. However, most previous studies focus on prompting the out-of-the-box LLMs, relying on their innate capability for collaboration, which may not improve LLMs' performance as shown recently. In this paper, we introduce a new post-training paradigm MAPoRL (Multi-Agent Post-co-training for collaborative LLMs with Reinforcement Learning), to explicitly elicit the collaborative behaviors and further unleash the power of multi-agentic LLM frameworks. In MAPoRL, multip"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.18439","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/2502.18439/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":"2502.18439","created_at":"2026-07-05T11:35:53.726688+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.18439v2","created_at":"2026-07-05T11:35:53.726688+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.18439","created_at":"2026-07-05T11:35:53.726688+00:00"},{"alias_kind":"pith_short_12","alias_value":"XWD5XKUG3EUV","created_at":"2026-07-05T11:35:53.726688+00:00"},{"alias_kind":"pith_short_16","alias_value":"XWD5XKUG3EUVTFDF","created_at":"2026-07-05T11:35:53.726688+00:00"},{"alias_kind":"pith_short_8","alias_value":"XWD5XKUG","created_at":"2026-07-05T11:35:53.726688+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.06935","citing_title":"Mathematical methods of reinforcement learning","ref_index":115,"is_internal_anchor":true},{"citing_arxiv_id":"2511.20857","citing_title":"Evo-Memory: Benchmarking LLM Agent Test-time Learning with Self-Evolving Memory","ref_index":73,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12943","citing_title":"Reinforced Collaboration in Multi-Agent Flow Networks","ref_index":25,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XWD5XKUG3EUVTFDFFWN4H66XMM","json":"https://pith.science/pith/XWD5XKUG3EUVTFDFFWN4H66XMM.json","graph_json":"https://pith.science/api/pith-number/XWD5XKUG3EUVTFDFFWN4H66XMM/graph.json","events_json":"https://pith.science/api/pith-number/XWD5XKUG3EUVTFDFFWN4H66XMM/events.json","paper":"https://pith.science/paper/XWD5XKUG"},"agent_actions":{"view_html":"https://pith.science/pith/XWD5XKUG3EUVTFDFFWN4H66XMM","download_json":"https://pith.science/pith/XWD5XKUG3EUVTFDFFWN4H66XMM.json","view_paper":"https://pith.science/paper/XWD5XKUG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.18439&json=true","fetch_graph":"https://pith.science/api/pith-number/XWD5XKUG3EUVTFDFFWN4H66XMM/graph.json","fetch_events":"https://pith.science/api/pith-number/XWD5XKUG3EUVTFDFFWN4H66XMM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XWD5XKUG3EUVTFDFFWN4H66XMM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XWD5XKUG3EUVTFDFFWN4H66XMM/action/storage_attestation","attest_author":"https://pith.science/pith/XWD5XKUG3EUVTFDFFWN4H66XMM/action/author_attestation","sign_citation":"https://pith.science/pith/XWD5XKUG3EUVTFDFFWN4H66XMM/action/citation_signature","submit_replication":"https://pith.science/pith/XWD5XKUG3EUVTFDFFWN4H66XMM/action/replication_record"}},"created_at":"2026-07-05T11:35:53.726688+00:00","updated_at":"2026-07-05T11:35:53.726688+00:00"}