{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:V76OY464JY4HTUOBU6SLE7PB5P","short_pith_number":"pith:V76OY464","schema_version":"1.0","canonical_sha256":"affcec73dc4e3879d1c1a7a4b27de1ebcf809402bf05512c73935a802a69112b","source":{"kind":"arxiv","id":"2504.12330","version":1},"attestation_state":"computed","paper":{"title":"HM-RAG: Hierarchical Multi-Agent Multimodal Retrieval Augmented Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ding Wang, Jun Ma, Junming Liu, Pei Liu, Ruoyu Yao, Siyuan Meng, Xin Liu","submitted_at":"2025-04-13T06:55:33Z","abstract_excerpt":"While Retrieval-Augmented Generation (RAG) augments Large Language Models (LLMs) with external knowledge, conventional single-agent RAG remains fundamentally limited in resolving complex queries demanding coordinated reasoning across heterogeneous data ecosystems. We present HM-RAG, a novel Hierarchical Multi-agent Multimodal RAG framework that pioneers collaborative intelligence for dynamic knowledge synthesis across structured, unstructured, and graph-based data. The framework is composed of three-tiered architecture with specialized agents: a Decomposition Agent that dissects complex querie"},"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":"2504.12330","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-13T06:55:33Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"72a868817f1175d21e6ddddf6570a2bf1e203aecffa9336094eb74bbd4a2d1a2","abstract_canon_sha256":"b4fdcc176fde8e74c5c2e7adecd9596631e8662b04394050ca02328452d5909c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:50:14.173916Z","signature_b64":"A1t+Eb/ryjIVpM0EFUwKyeh2ZIxYQJL4NsIel78RZY+ZlVCuHmMfsBGfKS4MvkHgPawHCkWIehqN84CyjoKVAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"affcec73dc4e3879d1c1a7a4b27de1ebcf809402bf05512c73935a802a69112b","last_reissued_at":"2026-07-05T10:50:14.173381Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:50:14.173381Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"HM-RAG: Hierarchical Multi-Agent Multimodal Retrieval Augmented Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ding Wang, Jun Ma, Junming Liu, Pei Liu, Ruoyu Yao, Siyuan Meng, Xin Liu","submitted_at":"2025-04-13T06:55:33Z","abstract_excerpt":"While Retrieval-Augmented Generation (RAG) augments Large Language Models (LLMs) with external knowledge, conventional single-agent RAG remains fundamentally limited in resolving complex queries demanding coordinated reasoning across heterogeneous data ecosystems. We present HM-RAG, a novel Hierarchical Multi-agent Multimodal RAG framework that pioneers collaborative intelligence for dynamic knowledge synthesis across structured, unstructured, and graph-based data. The framework is composed of three-tiered architecture with specialized agents: a Decomposition Agent that dissects complex querie"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.12330","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/2504.12330/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":"2504.12330","created_at":"2026-07-05T10:50:14.173448+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.12330v1","created_at":"2026-07-05T10:50:14.173448+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.12330","created_at":"2026-07-05T10:50:14.173448+00:00"},{"alias_kind":"pith_short_12","alias_value":"V76OY464JY4H","created_at":"2026-07-05T10:50:14.173448+00:00"},{"alias_kind":"pith_short_16","alias_value":"V76OY464JY4HTUOB","created_at":"2026-07-05T10:50:14.173448+00:00"},{"alias_kind":"pith_short_8","alias_value":"V76OY464","created_at":"2026-07-05T10:50:14.173448+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.27164","citing_title":"Query Symbolically or Retrieve Semantically? A Dataset and Method for Semi-Structured Question Answering","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2510.20505","citing_title":"RELOOP: Recursive Retrieval with Multi-Hop Reasoner and Planners for Heterogeneous QA","ref_index":19,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/V76OY464JY4HTUOBU6SLE7PB5P","json":"https://pith.science/pith/V76OY464JY4HTUOBU6SLE7PB5P.json","graph_json":"https://pith.science/api/pith-number/V76OY464JY4HTUOBU6SLE7PB5P/graph.json","events_json":"https://pith.science/api/pith-number/V76OY464JY4HTUOBU6SLE7PB5P/events.json","paper":"https://pith.science/paper/V76OY464"},"agent_actions":{"view_html":"https://pith.science/pith/V76OY464JY4HTUOBU6SLE7PB5P","download_json":"https://pith.science/pith/V76OY464JY4HTUOBU6SLE7PB5P.json","view_paper":"https://pith.science/paper/V76OY464","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.12330&json=true","fetch_graph":"https://pith.science/api/pith-number/V76OY464JY4HTUOBU6SLE7PB5P/graph.json","fetch_events":"https://pith.science/api/pith-number/V76OY464JY4HTUOBU6SLE7PB5P/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/V76OY464JY4HTUOBU6SLE7PB5P/action/timestamp_anchor","attest_storage":"https://pith.science/pith/V76OY464JY4HTUOBU6SLE7PB5P/action/storage_attestation","attest_author":"https://pith.science/pith/V76OY464JY4HTUOBU6SLE7PB5P/action/author_attestation","sign_citation":"https://pith.science/pith/V76OY464JY4HTUOBU6SLE7PB5P/action/citation_signature","submit_replication":"https://pith.science/pith/V76OY464JY4HTUOBU6SLE7PB5P/action/replication_record"}},"created_at":"2026-07-05T10:50:14.173448+00:00","updated_at":"2026-07-05T10:50:14.173448+00:00"}