{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:MRRGW5KUJXZNBSYHPDJSBPHCNR","short_pith_number":"pith:MRRGW5KU","schema_version":"1.0","canonical_sha256":"64626b75544df2d0cb0778d320bce26c45714011e782d0d151458bce497c731f","source":{"kind":"arxiv","id":"2508.05668","version":3},"attestation_state":"computed","paper":{"title":"A Survey of LLM-based Deep Search Agents: Paradigm, Optimization, Evaluation, and Challenges","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.IR","authors_text":"Jiachen Zhu, Jianghao Lin, Rong Shan, Te Gao, Weinan Zhang, Weiwen Liu, Yong Yu, Yongzhao Xiao, Yunjia Xi, Zheli Zhou","submitted_at":"2025-08-03T08:02:51Z","abstract_excerpt":"The advent of Large Language Models (LLMs) has significantly revolutionized web search. The emergence of LLM-based Search Agents marks a pivotal shift towards deeper, dynamic, autonomous information seeking. These agents can comprehend user intentions and environmental context and execute multi-turn retrieval with dynamic planning, extending search capabilities far beyond the web. Leading examples like OpenAI's Deep Research highlight their potential for deep information mining and real-world applications. This survey provides the first systematic analysis of search agents. We comprehensively "},"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":"2508.05668","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.IR","submitted_at":"2025-08-03T08:02:51Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"3b4728676485e93cce4bd17874f047629ac6982317e199dc8d5a0ac47d48735f","abstract_canon_sha256":"d7016a526a3f8ac6a64225448e69ea630aacb3031229162799eb8cd49033423e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:55:42.067543Z","signature_b64":"hFHGFTUA7dhQx+cINL5xcc3RgFe6qPyflww0qpihZejjfXJaRkrBjGRnjY853qAH1+WbpLXRykIa4teStUO2DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"64626b75544df2d0cb0778d320bce26c45714011e782d0d151458bce497c731f","last_reissued_at":"2026-07-05T11:55:42.067062Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:55:42.067062Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Survey of LLM-based Deep Search Agents: Paradigm, Optimization, Evaluation, and Challenges","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.IR","authors_text":"Jiachen Zhu, Jianghao Lin, Rong Shan, Te Gao, Weinan Zhang, Weiwen Liu, Yong Yu, Yongzhao Xiao, Yunjia Xi, Zheli Zhou","submitted_at":"2025-08-03T08:02:51Z","abstract_excerpt":"The advent of Large Language Models (LLMs) has significantly revolutionized web search. The emergence of LLM-based Search Agents marks a pivotal shift towards deeper, dynamic, autonomous information seeking. These agents can comprehend user intentions and environmental context and execute multi-turn retrieval with dynamic planning, extending search capabilities far beyond the web. Leading examples like OpenAI's Deep Research highlight their potential for deep information mining and real-world applications. This survey provides the first systematic analysis of search agents. We comprehensively "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.05668","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/2508.05668/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":"2508.05668","created_at":"2026-07-05T11:55:42.067118+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.05668v3","created_at":"2026-07-05T11:55:42.067118+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.05668","created_at":"2026-07-05T11:55:42.067118+00:00"},{"alias_kind":"pith_short_12","alias_value":"MRRGW5KUJXZN","created_at":"2026-07-05T11:55:42.067118+00:00"},{"alias_kind":"pith_short_16","alias_value":"MRRGW5KUJXZNBSYH","created_at":"2026-07-05T11:55:42.067118+00:00"},{"alias_kind":"pith_short_8","alias_value":"MRRGW5KU","created_at":"2026-07-05T11:55:42.067118+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":17,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.06374","citing_title":"VaseMuseum: Digital Intelligent Museum for Ancient Greek Pottery","ref_index":16,"is_internal_anchor":true},{"citing_arxiv_id":"2605.29555","citing_title":"From Blind Guess to Informed Judgment: Teaching LLMs to Evaluate Materials by Building Knowledge-Augmented Preference Signals","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2606.11926","citing_title":"Toward Generalist Autonomous Research via Hypothesis-Tree Refinement","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16986","citing_title":"Skills on the Fly: Test-Time Adaptive Skill Synthesis for LLM Agents","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27859","citing_title":"Rethinking Agentic Reinforcement Learning In Large Language Models","ref_index":103,"is_internal_anchor":false},{"citing_arxiv_id":"2510.00861","citing_title":"Erase to Improve: Erasable Reinforcement Learning for Search-Augmented LLMs","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2510.14703","citing_title":"ToolPRM: Fine-Grained Inference Scaling of Structured Outputs for Function Calling","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13534","citing_title":"Scaling Retrieval-Augmented Reasoning with Parallel Search and Explicit Merging","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04949","citing_title":"Learning to Retrieve from Agent Trajectories","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27859","citing_title":"Rethinking Agentic Reinforcement Learning In Large Language Models","ref_index":103,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10488","citing_title":"DeepRefine: Agent-Compiled Knowledge Refinement via Reinforcement Learning","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09915","citing_title":"Position: Academic Conferences are Potentially Facing Denominator Gaming Caused by Fully Automated Scientific Agents","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27859","citing_title":"Rethinking Agentic Reinforcement Learning In Large Language Models","ref_index":103,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00043","citing_title":"SiriusHelper: An LLM Agent-Based Operations Assistant for Big Data Platforms","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12890","citing_title":"Towards Long-horizon Agentic Multimodal Search","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08224","citing_title":"Externalization in LLM Agents: A Unified Review of Memory, Skills, Protocols and Harness Engineering","ref_index":162,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05952","citing_title":"Towards Trustworthy Report Generation: A Deep Research Agent with Progressive Confidence Estimation and Calibration","ref_index":23,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MRRGW5KUJXZNBSYHPDJSBPHCNR","json":"https://pith.science/pith/MRRGW5KUJXZNBSYHPDJSBPHCNR.json","graph_json":"https://pith.science/api/pith-number/MRRGW5KUJXZNBSYHPDJSBPHCNR/graph.json","events_json":"https://pith.science/api/pith-number/MRRGW5KUJXZNBSYHPDJSBPHCNR/events.json","paper":"https://pith.science/paper/MRRGW5KU"},"agent_actions":{"view_html":"https://pith.science/pith/MRRGW5KUJXZNBSYHPDJSBPHCNR","download_json":"https://pith.science/pith/MRRGW5KUJXZNBSYHPDJSBPHCNR.json","view_paper":"https://pith.science/paper/MRRGW5KU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.05668&json=true","fetch_graph":"https://pith.science/api/pith-number/MRRGW5KUJXZNBSYHPDJSBPHCNR/graph.json","fetch_events":"https://pith.science/api/pith-number/MRRGW5KUJXZNBSYHPDJSBPHCNR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MRRGW5KUJXZNBSYHPDJSBPHCNR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MRRGW5KUJXZNBSYHPDJSBPHCNR/action/storage_attestation","attest_author":"https://pith.science/pith/MRRGW5KUJXZNBSYHPDJSBPHCNR/action/author_attestation","sign_citation":"https://pith.science/pith/MRRGW5KUJXZNBSYHPDJSBPHCNR/action/citation_signature","submit_replication":"https://pith.science/pith/MRRGW5KUJXZNBSYHPDJSBPHCNR/action/replication_record"}},"created_at":"2026-07-05T11:55:42.067118+00:00","updated_at":"2026-07-05T11:55:42.067118+00:00"}