{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:VBRNBJUCDX4F7Z6VZGICCYIWKH","short_pith_number":"pith:VBRNBJUC","schema_version":"1.0","canonical_sha256":"a862d0a6821df85fe7d5c99021611651db7485d49ff08ad7e128ef1015727452","source":{"kind":"arxiv","id":"2403.10081","version":3},"attestation_state":"computed","paper":{"title":"DRAGIN: Dynamic Retrieval Augmented Generation based on the Information Needs of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.CL","authors_text":"Qingyao Ai, Weihang Su, Yichen Tang, Yiqun Liu, Zhijing Wu","submitted_at":"2024-03-15T07:45:37Z","abstract_excerpt":"Dynamic retrieval augmented generation (RAG) paradigm actively decides when and what to retrieve during the text generation process of Large Language Models (LLMs). There are two key elements of this paradigm: identifying the optimal moment to activate the retrieval module (deciding when to retrieve) and crafting the appropriate query once retrieval is triggered (determining what to retrieve). However, current dynamic RAG methods fall short in both aspects. Firstly, the strategies for deciding when to retrieve often rely on static rules. Moreover, the strategies for deciding what to retrieve t"},"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":"2403.10081","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-15T07:45:37Z","cross_cats_sorted":["cs.IR"],"title_canon_sha256":"884fa9147fb6d4ced78cd6cfd0f813ccfcf3e1b1c56e982d728bd854b1e9ee20","abstract_canon_sha256":"5ef98dbc4b9fc93450996dc5aec1b4864e27fdb28fc5a19a2138c284644b33fd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:09:57.371827Z","signature_b64":"+esSVrPjqaALAAtbXcwElBWibA8mZimGa3+C8chKUmAO2STVFP29vcWTEwpnZeotNQsV9h1rHWa9tDgvo2NxBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a862d0a6821df85fe7d5c99021611651db7485d49ff08ad7e128ef1015727452","last_reissued_at":"2026-07-05T09:09:57.371340Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:09:57.371340Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DRAGIN: Dynamic Retrieval Augmented Generation based on the Information Needs of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.CL","authors_text":"Qingyao Ai, Weihang Su, Yichen Tang, Yiqun Liu, Zhijing Wu","submitted_at":"2024-03-15T07:45:37Z","abstract_excerpt":"Dynamic retrieval augmented generation (RAG) paradigm actively decides when and what to retrieve during the text generation process of Large Language Models (LLMs). There are two key elements of this paradigm: identifying the optimal moment to activate the retrieval module (deciding when to retrieve) and crafting the appropriate query once retrieval is triggered (determining what to retrieve). However, current dynamic RAG methods fall short in both aspects. Firstly, the strategies for deciding when to retrieve often rely on static rules. Moreover, the strategies for deciding what to retrieve t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.10081","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/2403.10081/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":"2403.10081","created_at":"2026-07-05T09:09:57.371396+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.10081v3","created_at":"2026-07-05T09:09:57.371396+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.10081","created_at":"2026-07-05T09:09:57.371396+00:00"},{"alias_kind":"pith_short_12","alias_value":"VBRNBJUCDX4F","created_at":"2026-07-05T09:09:57.371396+00:00"},{"alias_kind":"pith_short_16","alias_value":"VBRNBJUCDX4F7Z6V","created_at":"2026-07-05T09:09:57.371396+00:00"},{"alias_kind":"pith_short_8","alias_value":"VBRNBJUC","created_at":"2026-07-05T09:09:57.371396+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":15,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.17449","citing_title":"MODE-RAG: Manifold Outlier Diagnosis and Energy-based Retrieval-Augmented Generation Evaluation","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2606.17041","citing_title":"MetaSyn: A Benchmark for LLM Agents on Meta-Analysis Articles from Nature Portfolio","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2606.17041","citing_title":"MetaSyn: A Benchmark for LLM Agents on Meta-Analysis Articles from Nature Portfolio","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2606.07299","citing_title":"DuMate-DeepResearch: An Auditable Multi-Agent System with Recursive Search and Rubric-Grounded Reasoning","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2606.17041","citing_title":"MetaSyn: A Benchmark for LLM Agents on Meta-Analysis Articles from Nature Portfolio","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2404.10981","citing_title":"A Survey on Retrieval-Augmented Text Generation for Large Language Models","ref_index":128,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18792","citing_title":"Trust or Abstain? A Self-Aware RAG Approach","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2510.11541","citing_title":"Question-Adaptive Graph Learning for Multi-hop Retrieval Augmented Generation","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2601.00376","citing_title":"In Line with Context: Repository-Level Code Generation via Context Inlining","ref_index":59,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05306","citing_title":"LLMs Should Express Uncertainty Explicitly","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2603.29002","citing_title":"Understand and Accelerate Memory Processing Pipeline for Large Language Model Inference","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2604.01413","citing_title":"Adaptive Stopping for Multi-Turn LLM Reasoning","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02501","citing_title":"ECG Foundation Models and Medical LLMs for Agentic Cardiovascular Intelligence at the Edge: A Review and Outlook","ref_index":151,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02011","citing_title":"Enhancing Judgment Document Generation via Agentic Legal Information Collection and Rubric-Guided Optimization","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05306","citing_title":"LLMs Should Express Uncertainty Explicitly","ref_index":8,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VBRNBJUCDX4F7Z6VZGICCYIWKH","json":"https://pith.science/pith/VBRNBJUCDX4F7Z6VZGICCYIWKH.json","graph_json":"https://pith.science/api/pith-number/VBRNBJUCDX4F7Z6VZGICCYIWKH/graph.json","events_json":"https://pith.science/api/pith-number/VBRNBJUCDX4F7Z6VZGICCYIWKH/events.json","paper":"https://pith.science/paper/VBRNBJUC"},"agent_actions":{"view_html":"https://pith.science/pith/VBRNBJUCDX4F7Z6VZGICCYIWKH","download_json":"https://pith.science/pith/VBRNBJUCDX4F7Z6VZGICCYIWKH.json","view_paper":"https://pith.science/paper/VBRNBJUC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.10081&json=true","fetch_graph":"https://pith.science/api/pith-number/VBRNBJUCDX4F7Z6VZGICCYIWKH/graph.json","fetch_events":"https://pith.science/api/pith-number/VBRNBJUCDX4F7Z6VZGICCYIWKH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VBRNBJUCDX4F7Z6VZGICCYIWKH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VBRNBJUCDX4F7Z6VZGICCYIWKH/action/storage_attestation","attest_author":"https://pith.science/pith/VBRNBJUCDX4F7Z6VZGICCYIWKH/action/author_attestation","sign_citation":"https://pith.science/pith/VBRNBJUCDX4F7Z6VZGICCYIWKH/action/citation_signature","submit_replication":"https://pith.science/pith/VBRNBJUCDX4F7Z6VZGICCYIWKH/action/replication_record"}},"created_at":"2026-07-05T09:09:57.371396+00:00","updated_at":"2026-07-05T09:09:57.371396+00:00"}