{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:GBLLO7PRKIS5DXOM37M3V7UUTV","short_pith_number":"pith:GBLLO7PR","schema_version":"1.0","canonical_sha256":"3056b77df15225d1ddccdfd9bafe949d5478851bea0a85e3885b6a805d44da78","source":{"kind":"arxiv","id":"2608.05967","version":1},"attestation_state":"computed","paper":{"title":"M$^3$Prune: Hierarchical Collaborative Pruning for Efficient Multi-Modal Multi-Agent Retrieval-Augmented Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.MM","authors_text":"Chen Chen, Chengyu Wang, Daiyang Yu, Taolin Zhang, Tingyuan Hu, Weizi shao, Xiaofeng He, Zijie Zhou","submitted_at":"2026-08-06T12:44:35Z","abstract_excerpt":"Recent advances in multi-modal retrieval-augmented generation (mRAG), which augments multi-modal large language models (MLLMs) with external knowledge, have shown that collective intelligence from multiple agents can outperform a single model through effective communication. Despite their strong performance, existing multi-agent systems incur substantial token overhead and computational cost, posing challenges for large-scale deployment. To address these issues, we propose a Multi-Modal Multi-agent hierarchical communication graph PRUNING framework, termed M3Prune. M3Prune eliminates redundant"},"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":"2608.05967","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.MM","submitted_at":"2026-08-06T12:44:35Z","cross_cats_sorted":[],"title_canon_sha256":"a8efb788b0eb8c7d228132a59b54baf201ccb6bdb520ff02bb9e7916741395ec","abstract_canon_sha256":"928a5108a36d67cc657ab194aecdb65199d7368923ca0a8aca968cb3deb83bed"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-07T00:54:45.023361Z","signature_b64":"c328N8UPU0JmL7JndPJL5dA1ig+21EoMSZM/C+64nkMUVj+pNB6jCUEUY+3GfqOk76/6B8RCF2SJjk8aWzyXDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3056b77df15225d1ddccdfd9bafe949d5478851bea0a85e3885b6a805d44da78","last_reissued_at":"2026-08-07T00:54:45.021860Z","signature_status":"signed_v1","first_computed_at":"2026-08-07T00:54:45.021860Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"M$^3$Prune: Hierarchical Collaborative Pruning for Efficient Multi-Modal Multi-Agent Retrieval-Augmented Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.MM","authors_text":"Chen Chen, Chengyu Wang, Daiyang Yu, Taolin Zhang, Tingyuan Hu, Weizi shao, Xiaofeng He, Zijie Zhou","submitted_at":"2026-08-06T12:44:35Z","abstract_excerpt":"Recent advances in multi-modal retrieval-augmented generation (mRAG), which augments multi-modal large language models (MLLMs) with external knowledge, have shown that collective intelligence from multiple agents can outperform a single model through effective communication. Despite their strong performance, existing multi-agent systems incur substantial token overhead and computational cost, posing challenges for large-scale deployment. To address these issues, we propose a Multi-Modal Multi-agent hierarchical communication graph PRUNING framework, termed M3Prune. M3Prune eliminates redundant"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.05967","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/2608.05967/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":"2608.05967","created_at":"2026-08-07T00:54:45.023431+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.05967v1","created_at":"2026-08-07T00:54:45.023431+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.05967","created_at":"2026-08-07T00:54:45.023431+00:00"},{"alias_kind":"pith_short_12","alias_value":"GBLLO7PRKIS5","created_at":"2026-08-07T00:54:45.023431+00:00"},{"alias_kind":"pith_short_16","alias_value":"GBLLO7PRKIS5DXOM","created_at":"2026-08-07T00:54:45.023431+00:00"},{"alias_kind":"pith_short_8","alias_value":"GBLLO7PR","created_at":"2026-08-07T00:54:45.023431+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GBLLO7PRKIS5DXOM37M3V7UUTV","json":"https://pith.science/pith/GBLLO7PRKIS5DXOM37M3V7UUTV.json","graph_json":"https://pith.science/api/pith-number/GBLLO7PRKIS5DXOM37M3V7UUTV/graph.json","events_json":"https://pith.science/api/pith-number/GBLLO7PRKIS5DXOM37M3V7UUTV/events.json","paper":"https://pith.science/paper/GBLLO7PR"},"agent_actions":{"view_html":"https://pith.science/pith/GBLLO7PRKIS5DXOM37M3V7UUTV","download_json":"https://pith.science/pith/GBLLO7PRKIS5DXOM37M3V7UUTV.json","view_paper":"https://pith.science/paper/GBLLO7PR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.05967&json=true","fetch_graph":"https://pith.science/api/pith-number/GBLLO7PRKIS5DXOM37M3V7UUTV/graph.json","fetch_events":"https://pith.science/api/pith-number/GBLLO7PRKIS5DXOM37M3V7UUTV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GBLLO7PRKIS5DXOM37M3V7UUTV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GBLLO7PRKIS5DXOM37M3V7UUTV/action/storage_attestation","attest_author":"https://pith.science/pith/GBLLO7PRKIS5DXOM37M3V7UUTV/action/author_attestation","sign_citation":"https://pith.science/pith/GBLLO7PRKIS5DXOM37M3V7UUTV/action/citation_signature","submit_replication":"https://pith.science/pith/GBLLO7PRKIS5DXOM37M3V7UUTV/action/replication_record"}},"created_at":"2026-08-07T00:54:45.023431+00:00","updated_at":"2026-08-07T00:54:45.023431+00:00"}