{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:7QOPFRMUJV4YAYI2GQVSRWPZPL","short_pith_number":"pith:7QOPFRMU","schema_version":"1.0","canonical_sha256":"fc1cf2c5944d7980611a342b28d9f97adcfd49fa589eb152524740c03f64b0c7","source":{"kind":"arxiv","id":"2505.16096","version":1},"attestation_state":"computed","paper":{"title":"Cosmos: A CXL-Based Full In-Memory System for Approximate Nearest Neighbor Search","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AR","authors_text":"Hyunjeong Shim, Jinin So, Jung Ho Ahn, Minyong Yoon, Sang-Soo Park, Seoyoung Ko, Si-Dong Roh, Sungmin Yun, Taeksang Song, Wanju Doh, Yongsuk Kwon","submitted_at":"2025-05-22T00:45:32Z","abstract_excerpt":"Retrieval-Augmented Generation (RAG) is crucial for improving the quality of large language models by injecting proper contexts extracted from external sources. RAG requires high-throughput, low-latency Approximate Nearest Neighbor Search (ANNS) over billion-scale vector databases. Conventional DRAM/SSD solutions face capacity/latency limits, whereas specialized hardware or RDMA clusters lack flexibility or incur network overhead. We present Cosmos, integrating general-purpose cores within CXL memory devices for full ANNS offload and introducing rank-level parallel distance computation to maxi"},"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":"2505.16096","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2025-05-22T00:45:32Z","cross_cats_sorted":[],"title_canon_sha256":"5ac8b34d153d1ec83164de70abdc0336fbca646526aaea1a0fbbbda4f0710a58","abstract_canon_sha256":"188bdb09922924182ee1834a9c82a9260db83e53dca6199cf2533561c86f3b01"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:07:29.875042Z","signature_b64":"hTODlv3qhcUC8/X6SiJnMNc9ppgoO7cqoav72uzUSBsFK7ETRvr8ygyQ2FGc4woprSiG5sWtCk8xyHGcbKrcDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fc1cf2c5944d7980611a342b28d9f97adcfd49fa589eb152524740c03f64b0c7","last_reissued_at":"2026-07-05T11:07:29.874634Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:07:29.874634Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Cosmos: A CXL-Based Full In-Memory System for Approximate Nearest Neighbor Search","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AR","authors_text":"Hyunjeong Shim, Jinin So, Jung Ho Ahn, Minyong Yoon, Sang-Soo Park, Seoyoung Ko, Si-Dong Roh, Sungmin Yun, Taeksang Song, Wanju Doh, Yongsuk Kwon","submitted_at":"2025-05-22T00:45:32Z","abstract_excerpt":"Retrieval-Augmented Generation (RAG) is crucial for improving the quality of large language models by injecting proper contexts extracted from external sources. RAG requires high-throughput, low-latency Approximate Nearest Neighbor Search (ANNS) over billion-scale vector databases. Conventional DRAM/SSD solutions face capacity/latency limits, whereas specialized hardware or RDMA clusters lack flexibility or incur network overhead. We present Cosmos, integrating general-purpose cores within CXL memory devices for full ANNS offload and introducing rank-level parallel distance computation to maxi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.16096","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/2505.16096/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":"2505.16096","created_at":"2026-07-05T11:07:29.874693+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.16096v1","created_at":"2026-07-05T11:07:29.874693+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.16096","created_at":"2026-07-05T11:07:29.874693+00:00"},{"alias_kind":"pith_short_12","alias_value":"7QOPFRMUJV4Y","created_at":"2026-07-05T11:07:29.874693+00:00"},{"alias_kind":"pith_short_16","alias_value":"7QOPFRMUJV4YAYI2","created_at":"2026-07-05T11:07:29.874693+00:00"},{"alias_kind":"pith_short_8","alias_value":"7QOPFRMU","created_at":"2026-07-05T11:07:29.874693+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/7QOPFRMUJV4YAYI2GQVSRWPZPL","json":"https://pith.science/pith/7QOPFRMUJV4YAYI2GQVSRWPZPL.json","graph_json":"https://pith.science/api/pith-number/7QOPFRMUJV4YAYI2GQVSRWPZPL/graph.json","events_json":"https://pith.science/api/pith-number/7QOPFRMUJV4YAYI2GQVSRWPZPL/events.json","paper":"https://pith.science/paper/7QOPFRMU"},"agent_actions":{"view_html":"https://pith.science/pith/7QOPFRMUJV4YAYI2GQVSRWPZPL","download_json":"https://pith.science/pith/7QOPFRMUJV4YAYI2GQVSRWPZPL.json","view_paper":"https://pith.science/paper/7QOPFRMU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.16096&json=true","fetch_graph":"https://pith.science/api/pith-number/7QOPFRMUJV4YAYI2GQVSRWPZPL/graph.json","fetch_events":"https://pith.science/api/pith-number/7QOPFRMUJV4YAYI2GQVSRWPZPL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7QOPFRMUJV4YAYI2GQVSRWPZPL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7QOPFRMUJV4YAYI2GQVSRWPZPL/action/storage_attestation","attest_author":"https://pith.science/pith/7QOPFRMUJV4YAYI2GQVSRWPZPL/action/author_attestation","sign_citation":"https://pith.science/pith/7QOPFRMUJV4YAYI2GQVSRWPZPL/action/citation_signature","submit_replication":"https://pith.science/pith/7QOPFRMUJV4YAYI2GQVSRWPZPL/action/replication_record"}},"created_at":"2026-07-05T11:07:29.874693+00:00","updated_at":"2026-07-05T11:07:29.874693+00:00"}