{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:WVAJGKVRISNBA6Q2SQ57BDB3V4","short_pith_number":"pith:WVAJGKVR","schema_version":"1.0","canonical_sha256":"b540932ab1449a107a1a943bf08c3baf3aedbec8a8f5bcbaf16d5908bbd2b198","source":{"kind":"arxiv","id":"2602.00471","version":2},"attestation_state":"computed","paper":{"title":"Dual Latent Memory for Visual Multi-agent System","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.AI","authors_text":"Bo Yin, Chengming Xu, Cheng Tan, Cheng Yang, Jiangning Zhang, Shuicheng Yan, Xiaobin Hu, Xinlei Yu, Yihao Hu, Yongbo He, Zhangquan Chen","submitted_at":"2026-01-31T02:49:10Z","abstract_excerpt":"While Visual Multi-Agent Systems (VMAS) promise to enhance comprehensive abilities through inter-agent collaboration, empirical evidence reveals a counter-intuitive \"scaling wall\": increasing agent turns often degrades performance while exponentially inflating token costs. We attribute this failure to the information bottleneck inherent in text-centric communication, where converting perceptual and thinking trajectories into discrete natural language inevitably induces semantic loss. To this end, we propose \\textbf{L}$\\mathbf{^{2}}$\\textbf{-VMAS}, a novel model-agnostic framework that enables "},"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":"2602.00471","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-01-31T02:49:10Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"f62cd07e97681fabc54476219728349dad8108dbdb6f48d19e2640fc2ecfe4b4","abstract_canon_sha256":"a8b7229a914fff09c033039c071f0e5dee81f882f26b400793dbae4b54ba4534"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-08T01:03:57.232577Z","signature_b64":"6feXaAd0BMxRcghMXd5ry2PfbOAFa4yCbYZ0BJUjMHYmeAbJR3TBNGBB4tQLHzQ/wOM1lvFakFPt/mfUHEjqAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b540932ab1449a107a1a943bf08c3baf3aedbec8a8f5bcbaf16d5908bbd2b198","last_reissued_at":"2026-06-08T01:03:57.231417Z","signature_status":"signed_v1","first_computed_at":"2026-06-08T01:03:57.231417Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Dual Latent Memory for Visual Multi-agent System","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.AI","authors_text":"Bo Yin, Chengming Xu, Cheng Tan, Cheng Yang, Jiangning Zhang, Shuicheng Yan, Xiaobin Hu, Xinlei Yu, Yihao Hu, Yongbo He, Zhangquan Chen","submitted_at":"2026-01-31T02:49:10Z","abstract_excerpt":"While Visual Multi-Agent Systems (VMAS) promise to enhance comprehensive abilities through inter-agent collaboration, empirical evidence reveals a counter-intuitive \"scaling wall\": increasing agent turns often degrades performance while exponentially inflating token costs. We attribute this failure to the information bottleneck inherent in text-centric communication, where converting perceptual and thinking trajectories into discrete natural language inevitably induces semantic loss. To this end, we propose \\textbf{L}$\\mathbf{^{2}}$\\textbf{-VMAS}, a novel model-agnostic framework that enables "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.00471","kind":"arxiv","version":2},"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/2602.00471/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":"2602.00471","created_at":"2026-06-08T01:03:57.231588+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.00471v2","created_at":"2026-06-08T01:03:57.231588+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.00471","created_at":"2026-06-08T01:03:57.231588+00:00"},{"alias_kind":"pith_short_12","alias_value":"WVAJGKVRISNB","created_at":"2026-06-08T01:03:57.231588+00:00"},{"alias_kind":"pith_short_16","alias_value":"WVAJGKVRISNBA6Q2","created_at":"2026-06-08T01:03:57.231588+00:00"},{"alias_kind":"pith_short_8","alias_value":"WVAJGKVR","created_at":"2026-06-08T01:03:57.231588+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":3,"sample":[{"citing_arxiv_id":"2605.18597","citing_title":"Latent Action Reparameterization for Efficient Agent Inference","ref_index":43,"is_internal_anchor":true},{"citing_arxiv_id":"2604.19516","citing_title":"From Experience to Skill: Multi-Agent Generative Engine Optimization via Reusable Strategy Learning","ref_index":88,"is_internal_anchor":true},{"citing_arxiv_id":"2604.17503","citing_title":"SkillGraph: Self-Evolving Multi-Agent Collaboration with Multimodal Graph Topology","ref_index":45,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WVAJGKVRISNBA6Q2SQ57BDB3V4","json":"https://pith.science/pith/WVAJGKVRISNBA6Q2SQ57BDB3V4.json","graph_json":"https://pith.science/api/pith-number/WVAJGKVRISNBA6Q2SQ57BDB3V4/graph.json","events_json":"https://pith.science/api/pith-number/WVAJGKVRISNBA6Q2SQ57BDB3V4/events.json","paper":"https://pith.science/paper/WVAJGKVR"},"agent_actions":{"view_html":"https://pith.science/pith/WVAJGKVRISNBA6Q2SQ57BDB3V4","download_json":"https://pith.science/pith/WVAJGKVRISNBA6Q2SQ57BDB3V4.json","view_paper":"https://pith.science/paper/WVAJGKVR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.00471&json=true","fetch_graph":"https://pith.science/api/pith-number/WVAJGKVRISNBA6Q2SQ57BDB3V4/graph.json","fetch_events":"https://pith.science/api/pith-number/WVAJGKVRISNBA6Q2SQ57BDB3V4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WVAJGKVRISNBA6Q2SQ57BDB3V4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WVAJGKVRISNBA6Q2SQ57BDB3V4/action/storage_attestation","attest_author":"https://pith.science/pith/WVAJGKVRISNBA6Q2SQ57BDB3V4/action/author_attestation","sign_citation":"https://pith.science/pith/WVAJGKVRISNBA6Q2SQ57BDB3V4/action/citation_signature","submit_replication":"https://pith.science/pith/WVAJGKVRISNBA6Q2SQ57BDB3V4/action/replication_record"}},"created_at":"2026-06-08T01:03:57.231588+00:00","updated_at":"2026-06-08T01:03:57.231588+00:00"}