{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:KBGI2EQLAMIW7NKFALRFG7RLKI","short_pith_number":"pith:KBGI2EQL","schema_version":"1.0","canonical_sha256":"504c8d120b03116fb54502e2537e2b522c0e6dfb26f52d05c9cd6c21619f9e79","source":{"kind":"arxiv","id":"2505.17005","version":1},"attestation_state":"computed","paper":{"title":"R1-Searcher++: Incentivizing the Dynamic Knowledge Acquisition of LLMs via Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.IR"],"primary_cat":"cs.CL","authors_text":"Huatong Song, Jiahao Zhao, Jinhao Jiang, Ji-Rong Wen, Lei Fang, Wayne Xin Zhao, Wenqing Tian, Yingqian Min, Yuhuan Wu, Zhipeng Chen","submitted_at":"2025-05-22T17:58:26Z","abstract_excerpt":"Large Language Models (LLMs) are powerful but prone to hallucinations due to static knowledge. Retrieval-Augmented Generation (RAG) helps by injecting external information, but current methods often are costly, generalize poorly, or ignore the internal knowledge of the model. In this paper, we introduce R1-Searcher++, a novel framework designed to train LLMs to adaptively leverage both internal and external knowledge sources. R1-Searcher++ employs a two-stage training strategy: an initial SFT Cold-start phase for preliminary format learning, followed by RL for Dynamic Knowledge Acquisition. Th"},"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":"2505.17005","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-22T17:58:26Z","cross_cats_sorted":["cs.AI","cs.IR"],"title_canon_sha256":"4c42d957265fcbf0c67a4d5ce543545bc81c5039cc4195359338f77b38fec956","abstract_canon_sha256":"36386db39b7ceecc8d2e093de54b7dd501c306e41a7735f7010002b33dd16db6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:07:44.428669Z","signature_b64":"l1LWlC3Ph3NItk4vB8NVXs8bKxccnHvQhQktIk9zK57bxkdowDfrIp6frkCeti+iU9DqX6plyejzTDhDgt4MAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"504c8d120b03116fb54502e2537e2b522c0e6dfb26f52d05c9cd6c21619f9e79","last_reissued_at":"2026-07-05T11:07:44.428147Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:07:44.428147Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"R1-Searcher++: Incentivizing the Dynamic Knowledge Acquisition of LLMs via Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.IR"],"primary_cat":"cs.CL","authors_text":"Huatong Song, Jiahao Zhao, Jinhao Jiang, Ji-Rong Wen, Lei Fang, Wayne Xin Zhao, Wenqing Tian, Yingqian Min, Yuhuan Wu, Zhipeng Chen","submitted_at":"2025-05-22T17:58:26Z","abstract_excerpt":"Large Language Models (LLMs) are powerful but prone to hallucinations due to static knowledge. Retrieval-Augmented Generation (RAG) helps by injecting external information, but current methods often are costly, generalize poorly, or ignore the internal knowledge of the model. In this paper, we introduce R1-Searcher++, a novel framework designed to train LLMs to adaptively leverage both internal and external knowledge sources. R1-Searcher++ employs a two-stage training strategy: an initial SFT Cold-start phase for preliminary format learning, followed by RL for Dynamic Knowledge Acquisition. Th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.17005","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/2505.17005/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":"2505.17005","created_at":"2026-07-05T11:07:44.428206+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.17005v1","created_at":"2026-07-05T11:07:44.428206+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.17005","created_at":"2026-07-05T11:07:44.428206+00:00"},{"alias_kind":"pith_short_12","alias_value":"KBGI2EQLAMIW","created_at":"2026-07-05T11:07:44.428206+00:00"},{"alias_kind":"pith_short_16","alias_value":"KBGI2EQLAMIW7NKF","created_at":"2026-07-05T11:07:44.428206+00:00"},{"alias_kind":"pith_short_8","alias_value":"KBGI2EQL","created_at":"2026-07-05T11:07:44.428206+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.24828","citing_title":"Test-Time Deep Thinking to Explore Implicit Rules","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2504.01990","citing_title":"Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systems","ref_index":83,"is_internal_anchor":false},{"citing_arxiv_id":"2509.02547","citing_title":"The Landscape of Agentic Reinforcement Learning for LLMs: A Survey","ref_index":279,"is_internal_anchor":false},{"citing_arxiv_id":"2510.11122","citing_title":"Learning to Trust: Dynamic Utilization of Retrieval-Augmented Generation for E-commerce Search Relevance","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2509.02544","citing_title":"UI-TARS-2 Technical Report: Advancing GUI Agent with Multi-Turn Reinforcement Learning","ref_index":62,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17337","citing_title":"AutoSearch: Adaptive Search Depth for Efficient Agentic RAG via Reinforcement Learning","ref_index":21,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KBGI2EQLAMIW7NKFALRFG7RLKI","json":"https://pith.science/pith/KBGI2EQLAMIW7NKFALRFG7RLKI.json","graph_json":"https://pith.science/api/pith-number/KBGI2EQLAMIW7NKFALRFG7RLKI/graph.json","events_json":"https://pith.science/api/pith-number/KBGI2EQLAMIW7NKFALRFG7RLKI/events.json","paper":"https://pith.science/paper/KBGI2EQL"},"agent_actions":{"view_html":"https://pith.science/pith/KBGI2EQLAMIW7NKFALRFG7RLKI","download_json":"https://pith.science/pith/KBGI2EQLAMIW7NKFALRFG7RLKI.json","view_paper":"https://pith.science/paper/KBGI2EQL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.17005&json=true","fetch_graph":"https://pith.science/api/pith-number/KBGI2EQLAMIW7NKFALRFG7RLKI/graph.json","fetch_events":"https://pith.science/api/pith-number/KBGI2EQLAMIW7NKFALRFG7RLKI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KBGI2EQLAMIW7NKFALRFG7RLKI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KBGI2EQLAMIW7NKFALRFG7RLKI/action/storage_attestation","attest_author":"https://pith.science/pith/KBGI2EQLAMIW7NKFALRFG7RLKI/action/author_attestation","sign_citation":"https://pith.science/pith/KBGI2EQLAMIW7NKFALRFG7RLKI/action/citation_signature","submit_replication":"https://pith.science/pith/KBGI2EQLAMIW7NKFALRFG7RLKI/action/replication_record"}},"created_at":"2026-07-05T11:07:44.428206+00:00","updated_at":"2026-07-05T11:07:44.428206+00:00"}