{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ZVOIMJCNRYHHBLFPF5HPNFLWQB","short_pith_number":"pith:ZVOIMJCN","schema_version":"1.0","canonical_sha256":"cd5c86244d8e0e70acaf2f4ef695768066de0eaf370d8340edb2f2e7fb945a8a","source":{"kind":"arxiv","id":"2406.05804","version":6},"attestation_state":"computed","paper":{"title":"A Review of Prominent Paradigms for LLM-Based Agents: Tool Use (Including RAG), Planning, and Feedback Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.SE"],"primary_cat":"cs.AI","authors_text":"Xinzhe Li","submitted_at":"2024-06-09T14:42:55Z","abstract_excerpt":"Tool use, planning, and feedback learning are currently three prominent paradigms for developing Large Language Model (LLM)-based agents across various tasks. Although numerous frameworks have been devised for each paradigm, their intricate workflows and inconsistent taxonomy create challenges in understanding and reviewing the frameworks across different paradigms. This survey introduces a unified taxonomy to systematically review and discuss these frameworks. Specifically, 1) the taxonomy defines environments/tasks, common LLM-profiled roles or LMPRs (policy models, evaluators, and dynamic m"},"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":"2406.05804","kind":"arxiv","version":6},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-06-09T14:42:55Z","cross_cats_sorted":["cs.CL","cs.SE"],"title_canon_sha256":"167e52e70cdada0b1c0d1b046efef0e7585c3df6e28387f2dfe5f4851f1cdbd8","abstract_canon_sha256":"75fc8d0e8c77f17b2c5a3551b12cac6e60421c70c805e1804a3db84939c75d8c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:42:26.316808Z","signature_b64":"ze82kGGBxkS6STIjoqeIxVGH93/lP1tczhJbQXomdMCAo4v1izjcAWJMTs70WEVg/Obg2o7lvFgv/jGmFRu3BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cd5c86244d8e0e70acaf2f4ef695768066de0eaf370d8340edb2f2e7fb945a8a","last_reissued_at":"2026-07-05T09:42:26.316261Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:42:26.316261Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Review of Prominent Paradigms for LLM-Based Agents: Tool Use (Including RAG), Planning, and Feedback Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.SE"],"primary_cat":"cs.AI","authors_text":"Xinzhe Li","submitted_at":"2024-06-09T14:42:55Z","abstract_excerpt":"Tool use, planning, and feedback learning are currently three prominent paradigms for developing Large Language Model (LLM)-based agents across various tasks. Although numerous frameworks have been devised for each paradigm, their intricate workflows and inconsistent taxonomy create challenges in understanding and reviewing the frameworks across different paradigms. This survey introduces a unified taxonomy to systematically review and discuss these frameworks. Specifically, 1) the taxonomy defines environments/tasks, common LLM-profiled roles or LMPRs (policy models, evaluators, and dynamic m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.05804","kind":"arxiv","version":6},"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/2406.05804/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":"2406.05804","created_at":"2026-07-05T09:42:26.316323+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.05804v6","created_at":"2026-07-05T09:42:26.316323+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.05804","created_at":"2026-07-05T09:42:26.316323+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZVOIMJCNRYHH","created_at":"2026-07-05T09:42:26.316323+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZVOIMJCNRYHHBLFP","created_at":"2026-07-05T09:42:26.316323+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZVOIMJCN","created_at":"2026-07-05T09:42:26.316323+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2503.21460","citing_title":"Large Language Model Agent: A Survey on Methodology, Applications and Challenges","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2506.20911","citing_title":"FaSTA$^*$: Fast-Slow Toolpath Agent with Subroutine Mining for Efficient Multi-turn Image Editing","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2604.01707","citing_title":"Memory in the LLM Era: Modular Architectures and Strategies in a Unified Framework","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06365","citing_title":"From Agent Loops to Deterministic Graphs: Execution Lineage for Reproducible AI-Native Work","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07339","citing_title":"Tools as Continuous Flow for Evolving Agentic Reasoning","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11465","citing_title":"Three Roles, One Model: Role Orchestration at Inference Time to Close the Performance Gap Between Small and Large Agents","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZVOIMJCNRYHHBLFPF5HPNFLWQB","json":"https://pith.science/pith/ZVOIMJCNRYHHBLFPF5HPNFLWQB.json","graph_json":"https://pith.science/api/pith-number/ZVOIMJCNRYHHBLFPF5HPNFLWQB/graph.json","events_json":"https://pith.science/api/pith-number/ZVOIMJCNRYHHBLFPF5HPNFLWQB/events.json","paper":"https://pith.science/paper/ZVOIMJCN"},"agent_actions":{"view_html":"https://pith.science/pith/ZVOIMJCNRYHHBLFPF5HPNFLWQB","download_json":"https://pith.science/pith/ZVOIMJCNRYHHBLFPF5HPNFLWQB.json","view_paper":"https://pith.science/paper/ZVOIMJCN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.05804&json=true","fetch_graph":"https://pith.science/api/pith-number/ZVOIMJCNRYHHBLFPF5HPNFLWQB/graph.json","fetch_events":"https://pith.science/api/pith-number/ZVOIMJCNRYHHBLFPF5HPNFLWQB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZVOIMJCNRYHHBLFPF5HPNFLWQB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZVOIMJCNRYHHBLFPF5HPNFLWQB/action/storage_attestation","attest_author":"https://pith.science/pith/ZVOIMJCNRYHHBLFPF5HPNFLWQB/action/author_attestation","sign_citation":"https://pith.science/pith/ZVOIMJCNRYHHBLFPF5HPNFLWQB/action/citation_signature","submit_replication":"https://pith.science/pith/ZVOIMJCNRYHHBLFPF5HPNFLWQB/action/replication_record"}},"created_at":"2026-07-05T09:42:26.316323+00:00","updated_at":"2026-07-05T09:42:26.316323+00:00"}