{"id":"37f8e026d264c2ebc22c0668028127bbd5f561fd16d33ee2c007ddf0ba00670f","title":"Petru Neague and coauthors present ZeroRAG for decentralized retrieval","argument":"Petru Neague and coauthors present ZeroRAG, a decentralized routing layer for retrieval-augmented generation (RAG), where documents are retrieved to provide context for a large language model (LLM). They target a problem in centralized RAG: organizations and individuals may be unable or unwilling to give a central party custody of their documents because of privacy, copyright or regulatory compliance. In ZeroRAG, documents stay with the peers that own them. Each peer keeps a bounded set of neighbor advertisements and a cache of routed queries. The system uses these to route queries through a best-first walk and progressively rewire the overlay toward semantically relevant neighbors. The authors say this self-organization improves recall over time without a central coordinator or global data structure. In evaluations with three real-world workloads and two decentralized baselines, they report better recall and fewer network messages than the baselines. They also report results approaching a centralized document store in recall and LLM answer accuracy, recovery after catastrophic peer failures, and scaling to 100 000 peers. Escape Meridian interpretation: the approach could matter where document owners want retrieval without handing over custody, though the abstract does not establish how its reported performance carries over to other settings.","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The evidence is the paper's abstract, which names three real-world workloads and two decentralized baselines but does not identify them or provide evaluation details. The abstract reports that ZeroRAG approaches a centralized document store in recall and LLM answer accuracy, but does not quantify that comparison.","speaker":{"name":"Neague, Petru, Stokkink, Quinten, de Vos, Martijn, Pires, Rafael, Randl, Mathis, Kermarrec, Anne-Marie, Pouwelse, Johan","roleAtSource":"Article author"},"source":{"title":"ZeroRAG: Adaptive Semantic Peer Routing for Decentralized Retrieval-Augmented Generation","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2611.07928","publishedAt":"2026-11-05","dateBasis":"Date basis not provided."},"evidence":[{"text":"The authors propose ZeroRAG as a decentralized routing layer for RAG that routes queries across peers without central orchestration or global data structures.","excerpt":"We present ZeroRAG, a decentralized routing layer for document retrieval in RAG that performs semantic query routing across an overlay of peers with- out global data structures or centralized orchestration.","url":"https://arxiv.org/abs/2611.07928","relationship":"Source-stated"},{"text":"In ZeroRAG, documents remain with their owners, while peers use local neighbor advertisements and a cache of routed queries to guide routing and rewire the overlay toward semantically relevant neighbors.","excerpt":"Documents remain with the peers that own them, and each peer keeps only lightweight local state: a bounded set of neighbor advertisements and a cache of recently routed queries.","url":"https://arxiv.org/abs/2611.07928","relationship":"Source-stated"},{"text":"The authors report that ZeroRAG outperforms two decentralized baselines on recall and network messages, approaches a centralized store's recall and LLM answer accuracy, self-heals after catastrophic peer failures, and scales to 100 000 peers.","excerpt":"We find that ZeroRAG outperforms baselines in terms of recall and required network messages, approaches the recall and LLM answer accuracy of a centralized document store, self-heals after catastrophic peer failures, and scales to 100 000 peers.","url":"https://arxiv.org/abs/2611.07928","relationship":"Source-stated"},{"text":"The approach could be relevant to document owners who want retrieval without handing custody of their documents to a central party; the abstract does not show how the results carry over to other settings.","excerpt":"documents are owned by organizations, devices, and individuals who cannot or will not surrender custody of them to a central party","url":"https://arxiv.org/abs/2611.07928","relationship":"Escape Meridian interpretation"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.07928","checkedAt":"2026-11-08T04:36:08.469Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"How do the reported results vary across the three real-world workloads and two decentralized baselines?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"Petru Neague and coauthors present ZeroRAG for decentralized retrieval","summary":"Petru Neague and coauthors present ZeroRAG, a decentralized routing layer for retrieval-augmented generation (RAG), where documents are retrieved to provide context for a large language model (LLM). They target a problem in centralized RAG: organizations and individuals may be unable or unwilling to give a central party custody of their documents because of privacy, copyright or regulatory compliance. In ZeroRAG, documents stay with the peers that own them. Each peer keeps a bounded set of neighbor advertisements and a cache of routed queries. The system uses these to route queries through a best-first walk and progressively rewire the overlay toward semantically relevant neighbors. The authors say this self-organization improves recall over time without a central coordinator or global data structure. In evaluations with three real-world workloads and two decentralized baselines, they report better recall and fewer network messages than the baselines. They also report results approaching a centralized document store in recall and LLM answer accuracy, recovery after catastrophic peer failures, and scaling to 100 000 peers. Escape Meridian interpretation: the approach could matter where document owners want retrieval without handing over custody, though the abstract does not establish how its reported performance carries over to other settings.","question":"How do the reported results vary across the three real-world workloads and two decentralized baselines?","claims":[{"text":"The authors propose ZeroRAG as a decentralized routing layer for RAG that routes queries across peers without central orchestration or global data structures.","excerpt":"We present ZeroRAG, a decentralized routing layer for document retrieval in RAG that performs semantic query routing across an overlay of peers with- out global data structures or centralized orchestration.","support":"source-stated"},{"text":"In ZeroRAG, documents remain with their owners, while peers use local neighbor advertisements and a cache of routed queries to guide routing and rewire the overlay toward semantically relevant neighbors.","excerpt":"Documents remain with the peers that own them, and each peer keeps only lightweight local state: a bounded set of neighbor advertisements and a cache of recently routed queries.","support":"source-stated"},{"text":"The authors report that ZeroRAG outperforms two decentralized baselines on recall and network messages, approaches a centralized store's recall and LLM answer accuracy, self-heals after catastrophic peer failures, and scales to 100 000 peers.","excerpt":"We find that ZeroRAG outperforms baselines in terms of recall and required network messages, approaches the recall and LLM answer accuracy of a centralized document store, self-heals after catastrophic peer failures, and scales to 100 000 peers.","support":"source-stated"},{"text":"The approach could be relevant to document owners who want retrieval without handing custody of their documents to a central party; the abstract does not show how the results carry over to other settings.","excerpt":"documents are owned by organizations, devices, and individuals who cannot or will not surrender custody of them to a central party","support":"interpretation"}],"limitations":["The evidence is the paper's abstract, which names three real-world workloads and two decentralized baselines but does not identify them or provide evaluation details.","The abstract reports that ZeroRAG approaches a centralized document store in recall and LLM answer accuracy, but does not quantify that comparison."],"mediaId":"87f999593859450e492d2c84ef9491222ad5ba56ced27c740a217a8fc2e81162","correctionReason":null,"id":"37f8e026d264c2ebc22c0668028127bbd5f561fd16d33ee2c007ddf0ba00670f","publicationId":"a68158c0-24ea-4124-8dd8-221d15bee6b7","version":1,"sourceUrl":"https://arxiv.org/abs/2611.07928","sourceTitle":"ZeroRAG: Adaptive Semantic Peer Routing for Decentralized Retrieval-Augmented Generation","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-11-05","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-11-06T05:03:23.125Z","createdAt":"2026-11-08T03:32:25.433Z","historical":true,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"87f999593859450e492d2c84ef9491222ad5ba56ced27c740a217a8fc2e81162","kind":"article","url":"https://arxiv.org/abs/2611.07928","sourceUrl":"https://arxiv.org/abs/2611.07928","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-11-08T03:18:45.161Z","method":"acquired_source","contentType":"text/html"},"recommendationReason":"Recommended by Jev for understanding this source; checked by Luna."}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.07928","checkedAt":"2026-11-08T04:36:08.469Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v5","mediaId":"87f999593859450e492d2c84ef9491222ad5ba56ced27c740a217a8fc2e81162","confidence":0.79,"provider":"typesafe","model":"jev-1.13.0","inputHash":"10dbff0ad5f3c788811ac920f73572fa6122c5681886e1e11aa4d0c7bdb8c139","decidedAt":"2026-11-08T03:18:45.894Z"},"provenance":{"textHash":"45df1f0a7df7535f08e22c50812f1d601a75c8a4dc797fda4fcb26d115f29847","observationId":"286662709c1bfa96d67fcec5a36b99b290e644e9d086a616bbda22713da63226","retrievedAt":"2026-11-08T03:18:45.045Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"322bf1f2-6d61-470f-88f2-383c4f983240"},"reviewer":{"model":"gpt-6-luna","runId":"bca0bb2e-aed8-46ce-87e0-4df7306197c3"}},"audit":{"status":"passed","reviewedAt":"2026-11-08T03:32:25.433Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Neague, Petru","url":null,"kind":"unknown"},{"name":"Stokkink, Quinten","url":null,"kind":"unknown"},{"name":"de Vos, Martijn","url":null,"kind":"unknown"},{"name":"Pires, Rafael","url":null,"kind":"unknown"},{"name":"Randl, Mathis","url":null,"kind":"unknown"},{"name":"Kermarrec, Anne-Marie","url":null,"kind":"unknown"},{"name":"Pouwelse, Johan","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Neague, Petru","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Neague, Petru","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.07928","locator":"head/meta[name=author|citation_author]"},"assertionId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-08T04:36:08.469Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12","sourceUrl":"https://arxiv.org/abs/2611.07928","contentSha256":"eb10f9056cc1894363859bcbb052ea1ea44e217e9cd8e469a6983bf42a30dcbe","textHash":"45df1f0a7df7535f08e22c50812f1d601a75c8a4dc797fda4fcb26d115f29847","observedAt":"2026-11-08T04:36:08.469Z"}},{"actor":{"name":"Stokkink, Quinten","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Stokkink, Quinten","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.07928","locator":"head/meta[name=author|citation_author]"},"assertionId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12:participation:1","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-08T04:36:08.469Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12","sourceUrl":"https://arxiv.org/abs/2611.07928","contentSha256":"eb10f9056cc1894363859bcbb052ea1ea44e217e9cd8e469a6983bf42a30dcbe","textHash":"45df1f0a7df7535f08e22c50812f1d601a75c8a4dc797fda4fcb26d115f29847","observedAt":"2026-11-08T04:36:08.469Z"}},{"actor":{"name":"de Vos, Martijn","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"de Vos, Martijn","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.07928","locator":"head/meta[name=author|citation_author]"},"assertionId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12:participation:2","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-08T04:36:08.469Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12","sourceUrl":"https://arxiv.org/abs/2611.07928","contentSha256":"eb10f9056cc1894363859bcbb052ea1ea44e217e9cd8e469a6983bf42a30dcbe","textHash":"45df1f0a7df7535f08e22c50812f1d601a75c8a4dc797fda4fcb26d115f29847","observedAt":"2026-11-08T04:36:08.469Z"}},{"actor":{"name":"Pires, Rafael","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Pires, Rafael","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.07928","locator":"head/meta[name=author|citation_author]"},"assertionId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12:participation:3","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-08T04:36:08.469Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12","sourceUrl":"https://arxiv.org/abs/2611.07928","contentSha256":"eb10f9056cc1894363859bcbb052ea1ea44e217e9cd8e469a6983bf42a30dcbe","textHash":"45df1f0a7df7535f08e22c50812f1d601a75c8a4dc797fda4fcb26d115f29847","observedAt":"2026-11-08T04:36:08.469Z"}},{"actor":{"name":"Randl, Mathis","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Randl, Mathis","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.07928","locator":"head/meta[name=author|citation_author]"},"assertionId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12:participation:4","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-08T04:36:08.469Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12","sourceUrl":"https://arxiv.org/abs/2611.07928","contentSha256":"eb10f9056cc1894363859bcbb052ea1ea44e217e9cd8e469a6983bf42a30dcbe","textHash":"45df1f0a7df7535f08e22c50812f1d601a75c8a4dc797fda4fcb26d115f29847","observedAt":"2026-11-08T04:36:08.469Z"}},{"actor":{"name":"Kermarrec, Anne-Marie","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Kermarrec, Anne-Marie","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.07928","locator":"head/meta[name=author|citation_author]"},"assertionId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12:participation:5","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-08T04:36:08.469Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12","sourceUrl":"https://arxiv.org/abs/2611.07928","contentSha256":"eb10f9056cc1894363859bcbb052ea1ea44e217e9cd8e469a6983bf42a30dcbe","textHash":"45df1f0a7df7535f08e22c50812f1d601a75c8a4dc797fda4fcb26d115f29847","observedAt":"2026-11-08T04:36:08.469Z"}},{"actor":{"name":"Pouwelse, Johan","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Pouwelse, Johan","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.07928","locator":"head/meta[name=author|citation_author]"},"assertionId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12:participation:6","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-08T04:36:08.469Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12","sourceUrl":"https://arxiv.org/abs/2611.07928","contentSha256":"eb10f9056cc1894363859bcbb052ea1ea44e217e9cd8e469a6983bf42a30dcbe","textHash":"45df1f0a7df7535f08e22c50812f1d601a75c8a4dc797fda4fcb26d115f29847","observedAt":"2026-11-08T04:36:08.469Z"}}],"participationCoverage":{"total":7,"returned":7,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2611.07928","snapshotId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12","contentSha256":"eb10f9056cc1894363859bcbb052ea1ea44e217e9cd8e469a6983bf42a30dcbe","textHash":"45df1f0a7df7535f08e22c50812f1d601a75c8a4dc797fda4fcb26d115f29847","observedAt":"2026-11-08T04:36:08.469Z","basis":"publisher-page-metadata","sourceLinks":49,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/37f8e026d264c2ebc22c0668028127bbd5f561fd16d33ee2c007ddf0ba00670f/source"}},"canonicalUrl":"https://escapemeridian.com/readings/petru-neague-and-coauthors-present-zerorag-for-decentralized-37f8e026","formats":{"html":"/readings/petru-neague-and-coauthors-present-zerorag-for-decentralized-37f8e026","markdown":"/readings/petru-neague-and-coauthors-present-zerorag-for-decentralized-37f8e026.md","json":"/readings/petru-neague-and-coauthors-present-zerorag-for-decentralized-37f8e026.json"},"attribution":"Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims.","maintainedGuideLinks":[],"readerContext":{"version":"reader-context-v1","attribution":"Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.","terms":[{"id":"term-rag","name":"Retrieval-augmented generation","explanation":"A setup in which a model first looks up relevant documents and then answers using them.","referenceUrl":"https://developers.google.com/machine-learning/glossary#retrieval-augmented-generation-rag"},{"id":"term-llm","name":"Large language model","explanation":"A language model with many learned numerical settings, used to process and generate language.","referenceUrl":"https://developers.google.com/machine-learning/glossary#large-language-model"}],"entities":[{"id":"arxiv","name":"arXiv","kind":"organization","context":"arXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed.","referenceUrls":["https://arxiv.org/","https://info.arxiv.org/help/api/index.html"],"path":"/entities/arxiv","json":"/entities/arxiv.json","markdown":"/entities/arxiv.md"},{"id":"time","name":"TIME","kind":"organization","context":"TIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services.","referenceUrls":["https://time.com/"],"path":"/entities/time","json":"/entities/time.json","markdown":"/entities/time.md"}]},"researchQuestions":[],"withdrawn":false,"references":{"schemaVersion":1,"target":{"type":"reading","id":"37f8e026d264c2ebc22c0668028127bbd5f561fd16d33ee2c007ddf0ba00670f","revision":"a68158c0-24ea-4124-8dd8-221d15bee6b7","revisionKind":"publication-id","title":"Petru Neague and coauthors present ZeroRAG for decentralized retrieval","url":"https://escapemeridian.com/readings/petru-neague-and-coauthors-present-zerorag-for-decentralized-37f8e026"},"source":{"type":"source","id":"https://arxiv.org/abs/2611.07928","revision":"45df1f0a7df7535f08e22c50812f1d601a75c8a4dc797fda4fcb26d115f29847","revisionKind":"sha256-source-text","title":"ZeroRAG: Adaptive Semantic Peer Routing for Decentralized Retrieval-Augmented Generation","url":"https://arxiv.org/abs/2611.07928"},"status":"available","items":[],"next":null,"scope":"Explicit citations in public readings and public file attachments to this reading. Shared topics are not citations. Other artifact types need an explicit reference adapter."},"revisions":[{"id":"37f8e026d264c2ebc22c0668028127bbd5f561fd16d33ee2c007ddf0ba00670f","title":"Petru Neague and coauthors present ZeroRAG for decentralized retrieval","argument":"Petru Neague and coauthors present ZeroRAG, a decentralized routing layer for retrieval-augmented generation (RAG), where documents are retrieved to provide context for a large language model (LLM). They target a problem in centralized RAG: organizations and individuals may be unable or unwilling to give a central party custody of their documents because of privacy, copyright or regulatory compliance. In ZeroRAG, documents stay with the peers that own them. Each peer keeps a bounded set of neighbor advertisements and a cache of routed queries. The system uses these to route queries through a best-first walk and progressively rewire the overlay toward semantically relevant neighbors. The authors say this self-organization improves recall over time without a central coordinator or global data structure. In evaluations with three real-world workloads and two decentralized baselines, they report better recall and fewer network messages than the baselines. They also report results approaching a centralized document store in recall and LLM answer accuracy, recovery after catastrophic peer failures, and scaling to 100 000 peers. Escape Meridian interpretation: the approach could matter where document owners want retrieval without handing over custody, though the abstract does not establish how its reported performance carries over to other settings.","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The evidence is the paper's abstract, which names three real-world workloads and two decentralized baselines but does not identify them or provide evaluation details. The abstract reports that ZeroRAG approaches a centralized document store in recall and LLM answer accuracy, but does not quantify that comparison.","speaker":{"name":"Neague, Petru, Stokkink, Quinten, de Vos, Martijn, Pires, Rafael, Randl, Mathis, Kermarrec, Anne-Marie, Pouwelse, Johan","roleAtSource":"Article author"},"source":{"title":"ZeroRAG: Adaptive Semantic Peer Routing for Decentralized Retrieval-Augmented Generation","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2611.07928","publishedAt":"2026-11-05","dateBasis":"Date basis not provided."},"evidence":[{"text":"The authors propose ZeroRAG as a decentralized routing layer for RAG that routes queries across peers without central orchestration or global data structures.","excerpt":"We present ZeroRAG, a decentralized routing layer for document retrieval in RAG that performs semantic query routing across an overlay of peers with- out global data structures or centralized orchestration.","url":"https://arxiv.org/abs/2611.07928","relationship":"Source-stated"},{"text":"In ZeroRAG, documents remain with their owners, while peers use local neighbor advertisements and a cache of routed queries to guide routing and rewire the overlay toward semantically relevant neighbors.","excerpt":"Documents remain with the peers that own them, and each peer keeps only lightweight local state: a bounded set of neighbor advertisements and a cache of recently routed queries.","url":"https://arxiv.org/abs/2611.07928","relationship":"Source-stated"},{"text":"The authors report that ZeroRAG outperforms two decentralized baselines on recall and network messages, approaches a centralized store's recall and LLM answer accuracy, self-heals after catastrophic peer failures, and scales to 100 000 peers.","excerpt":"We find that ZeroRAG outperforms baselines in terms of recall and required network messages, approaches the recall and LLM answer accuracy of a centralized document store, self-heals after catastrophic peer failures, and scales to 100 000 peers.","url":"https://arxiv.org/abs/2611.07928","relationship":"Source-stated"},{"text":"The approach could be relevant to document owners who want retrieval without handing custody of their documents to a central party; the abstract does not show how the results carry over to other settings.","excerpt":"documents are owned by organizations, devices, and individuals who cannot or will not surrender custody of them to a central party","url":"https://arxiv.org/abs/2611.07928","relationship":"Escape Meridian interpretation"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.07928","checkedAt":"2026-11-08T04:36:08.469Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"How do the reported results vary across the three real-world workloads and two decentralized baselines?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"Petru Neague and coauthors present ZeroRAG for decentralized retrieval","summary":"Petru Neague and coauthors present ZeroRAG, a decentralized routing layer for retrieval-augmented generation (RAG), where documents are retrieved to provide context for a large language model (LLM). They target a problem in centralized RAG: organizations and individuals may be unable or unwilling to give a central party custody of their documents because of privacy, copyright or regulatory compliance. In ZeroRAG, documents stay with the peers that own them. Each peer keeps a bounded set of neighbor advertisements and a cache of routed queries. The system uses these to route queries through a best-first walk and progressively rewire the overlay toward semantically relevant neighbors. The authors say this self-organization improves recall over time without a central coordinator or global data structure. In evaluations with three real-world workloads and two decentralized baselines, they report better recall and fewer network messages than the baselines. They also report results approaching a centralized document store in recall and LLM answer accuracy, recovery after catastrophic peer failures, and scaling to 100 000 peers. Escape Meridian interpretation: the approach could matter where document owners want retrieval without handing over custody, though the abstract does not establish how its reported performance carries over to other settings.","question":"How do the reported results vary across the three real-world workloads and two decentralized baselines?","claims":[{"text":"The authors propose ZeroRAG as a decentralized routing layer for RAG that routes queries across peers without central orchestration or global data structures.","excerpt":"We present ZeroRAG, a decentralized routing layer for document retrieval in RAG that performs semantic query routing across an overlay of peers with- out global data structures or centralized orchestration.","support":"source-stated"},{"text":"In ZeroRAG, documents remain with their owners, while peers use local neighbor advertisements and a cache of routed queries to guide routing and rewire the overlay toward semantically relevant neighbors.","excerpt":"Documents remain with the peers that own them, and each peer keeps only lightweight local state: a bounded set of neighbor advertisements and a cache of recently routed queries.","support":"source-stated"},{"text":"The authors report that ZeroRAG outperforms two decentralized baselines on recall and network messages, approaches a centralized store's recall and LLM answer accuracy, self-heals after catastrophic peer failures, and scales to 100 000 peers.","excerpt":"We find that ZeroRAG outperforms baselines in terms of recall and required network messages, approaches the recall and LLM answer accuracy of a centralized document store, self-heals after catastrophic peer failures, and scales to 100 000 peers.","support":"source-stated"},{"text":"The approach could be relevant to document owners who want retrieval without handing custody of their documents to a central party; the abstract does not show how the results carry over to other settings.","excerpt":"documents are owned by organizations, devices, and individuals who cannot or will not surrender custody of them to a central party","support":"interpretation"}],"limitations":["The evidence is the paper's abstract, which names three real-world workloads and two decentralized baselines but does not identify them or provide evaluation details.","The abstract reports that ZeroRAG approaches a centralized document store in recall and LLM answer accuracy, but does not quantify that comparison."],"mediaId":"87f999593859450e492d2c84ef9491222ad5ba56ced27c740a217a8fc2e81162","correctionReason":null,"id":"37f8e026d264c2ebc22c0668028127bbd5f561fd16d33ee2c007ddf0ba00670f","publicationId":"a68158c0-24ea-4124-8dd8-221d15bee6b7","version":1,"sourceUrl":"https://arxiv.org/abs/2611.07928","sourceTitle":"ZeroRAG: Adaptive Semantic Peer Routing for Decentralized Retrieval-Augmented Generation","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-11-05","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-11-06T05:03:23.125Z","createdAt":"2026-11-08T03:32:25.433Z","historical":true,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"87f999593859450e492d2c84ef9491222ad5ba56ced27c740a217a8fc2e81162","kind":"article","url":"https://arxiv.org/abs/2611.07928","sourceUrl":"https://arxiv.org/abs/2611.07928","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-11-08T03:18:45.161Z","method":"acquired_source","contentType":"text/html"},"recommendationReason":"Recommended by Jev for understanding this source; checked by Luna."}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.07928","checkedAt":"2026-11-08T04:36:08.469Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v5","mediaId":"87f999593859450e492d2c84ef9491222ad5ba56ced27c740a217a8fc2e81162","confidence":0.79,"provider":"typesafe","model":"jev-1.13.0","inputHash":"10dbff0ad5f3c788811ac920f73572fa6122c5681886e1e11aa4d0c7bdb8c139","decidedAt":"2026-11-08T03:18:45.894Z"},"provenance":{"textHash":"45df1f0a7df7535f08e22c50812f1d601a75c8a4dc797fda4fcb26d115f29847","observationId":"286662709c1bfa96d67fcec5a36b99b290e644e9d086a616bbda22713da63226","retrievedAt":"2026-11-08T03:18:45.045Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"322bf1f2-6d61-470f-88f2-383c4f983240"},"reviewer":{"model":"gpt-6-luna","runId":"bca0bb2e-aed8-46ce-87e0-4df7306197c3"}},"audit":{"status":"passed","reviewedAt":"2026-11-08T03:32:25.433Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Neague, Petru","url":null,"kind":"unknown"},{"name":"Stokkink, Quinten","url":null,"kind":"unknown"},{"name":"de Vos, Martijn","url":null,"kind":"unknown"},{"name":"Pires, Rafael","url":null,"kind":"unknown"},{"name":"Randl, Mathis","url":null,"kind":"unknown"},{"name":"Kermarrec, Anne-Marie","url":null,"kind":"unknown"},{"name":"Pouwelse, Johan","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Neague, Petru","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Neague, Petru","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.07928","locator":"head/meta[name=author|citation_author]"},"assertionId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-08T04:36:08.469Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12","sourceUrl":"https://arxiv.org/abs/2611.07928","contentSha256":"eb10f9056cc1894363859bcbb052ea1ea44e217e9cd8e469a6983bf42a30dcbe","textHash":"45df1f0a7df7535f08e22c50812f1d601a75c8a4dc797fda4fcb26d115f29847","observedAt":"2026-11-08T04:36:08.469Z"}},{"actor":{"name":"Stokkink, Quinten","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Stokkink, Quinten","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.07928","locator":"head/meta[name=author|citation_author]"},"assertionId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12:participation:1","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-08T04:36:08.469Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12","sourceUrl":"https://arxiv.org/abs/2611.07928","contentSha256":"eb10f9056cc1894363859bcbb052ea1ea44e217e9cd8e469a6983bf42a30dcbe","textHash":"45df1f0a7df7535f08e22c50812f1d601a75c8a4dc797fda4fcb26d115f29847","observedAt":"2026-11-08T04:36:08.469Z"}},{"actor":{"name":"de Vos, Martijn","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"de Vos, Martijn","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.07928","locator":"head/meta[name=author|citation_author]"},"assertionId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12:participation:2","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-08T04:36:08.469Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12","sourceUrl":"https://arxiv.org/abs/2611.07928","contentSha256":"eb10f9056cc1894363859bcbb052ea1ea44e217e9cd8e469a6983bf42a30dcbe","textHash":"45df1f0a7df7535f08e22c50812f1d601a75c8a4dc797fda4fcb26d115f29847","observedAt":"2026-11-08T04:36:08.469Z"}},{"actor":{"name":"Pires, Rafael","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Pires, Rafael","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.07928","locator":"head/meta[name=author|citation_author]"},"assertionId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12:participation:3","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-08T04:36:08.469Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12","sourceUrl":"https://arxiv.org/abs/2611.07928","contentSha256":"eb10f9056cc1894363859bcbb052ea1ea44e217e9cd8e469a6983bf42a30dcbe","textHash":"45df1f0a7df7535f08e22c50812f1d601a75c8a4dc797fda4fcb26d115f29847","observedAt":"2026-11-08T04:36:08.469Z"}},{"actor":{"name":"Randl, Mathis","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Randl, Mathis","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.07928","locator":"head/meta[name=author|citation_author]"},"assertionId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12:participation:4","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-08T04:36:08.469Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12","sourceUrl":"https://arxiv.org/abs/2611.07928","contentSha256":"eb10f9056cc1894363859bcbb052ea1ea44e217e9cd8e469a6983bf42a30dcbe","textHash":"45df1f0a7df7535f08e22c50812f1d601a75c8a4dc797fda4fcb26d115f29847","observedAt":"2026-11-08T04:36:08.469Z"}},{"actor":{"name":"Kermarrec, Anne-Marie","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Kermarrec, Anne-Marie","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.07928","locator":"head/meta[name=author|citation_author]"},"assertionId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12:participation:5","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-08T04:36:08.469Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12","sourceUrl":"https://arxiv.org/abs/2611.07928","contentSha256":"eb10f9056cc1894363859bcbb052ea1ea44e217e9cd8e469a6983bf42a30dcbe","textHash":"45df1f0a7df7535f08e22c50812f1d601a75c8a4dc797fda4fcb26d115f29847","observedAt":"2026-11-08T04:36:08.469Z"}},{"actor":{"name":"Pouwelse, Johan","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Pouwelse, Johan","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.07928","locator":"head/meta[name=author|citation_author]"},"assertionId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12:participation:6","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-08T04:36:08.469Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12","sourceUrl":"https://arxiv.org/abs/2611.07928","contentSha256":"eb10f9056cc1894363859bcbb052ea1ea44e217e9cd8e469a6983bf42a30dcbe","textHash":"45df1f0a7df7535f08e22c50812f1d601a75c8a4dc797fda4fcb26d115f29847","observedAt":"2026-11-08T04:36:08.469Z"}}],"participationCoverage":{"total":7,"returned":7,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2611.07928","snapshotId":"5600bd70284ba104960a0568d6ce146ce95870f3dd6ef24092260574b2a42f12","contentSha256":"eb10f9056cc1894363859bcbb052ea1ea44e217e9cd8e469a6983bf42a30dcbe","textHash":"45df1f0a7df7535f08e22c50812f1d601a75c8a4dc797fda4fcb26d115f29847","observedAt":"2026-11-08T04:36:08.469Z","basis":"publisher-page-metadata","sourceLinks":49,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/37f8e026d264c2ebc22c0668028127bbd5f561fd16d33ee2c007ddf0ba00670f/source"}}}],"audits":[{"publicationId":"a68158c0-24ea-4124-8dd8-221d15bee6b7","verdict":"passed","findings":[{"dimension":"attribution","status":"pass","explanation":"The draft correctly attributes ZeroRAG to Petru Neague and coauthors, who are listed as authors. It keeps the performance results framed as the authors' reported findings, not established facts.","excerpt":"Authors:Petru Neague, Quinten Stokkink, Martijn de Vos, Rafael Pires, Mathis Randl, Anne-Marie Kermarrec, Johan Pouwelse"},{"dimension":"summary","status":"pass","explanation":"The draft leads with the actor and the development. It explains RAG and LLM in plain terms and states the motivating custody problem: privacy, copyright and compliance. It also preserves the mechanism and the reported results. Terms such as overlay, best-first walk and recall are kept exactly and suit term-library tooltips. The plain gloss that documents stay with their owners keeps the core idea readable. The title is accurate, and the question is a genuine open issue about the evaluation details.","excerpt":"Documents remain with the peers that own them, and each peer keeps only lightweight local state"},{"dimension":"date","status":"pass","explanation":"The original publication date of 2026-11-05 matches the v1 submission date on arXiv. The feed date of 2026-11-06 is correctly not used.","excerpt":"[Submitted on 5 Nov 2026]"},{"dimension":"support","status":"pass","explanation":"Each claim matches the abstract. The interpretation is labeled as Escape Meridian's, is conditional, carries an excerpt and keeps the gap: it notes the abstract does not establish generalization. The limitations accurately say the workloads and baselines are unnamed and the centralized comparison is unquantified. No invented beneficiaries or costs appear.","excerpt":"approaches the recall and LLM answer accuracy of a centralized document store, self-heals after catastrophic peer failures, and scales to 100 000 peers"},{"dimension":"media","status":"pass","explanation":"The selected media is the verified direct arXiv abstract page for this paper. It is the most direct access to the source.","excerpt":null}],"createdAt":"2026-11-08T03:32:25.433Z","model":"anthropic/claude-opus-5.5"}],"correctionUrl":"https://escapemeridian.com/corrections/"}