{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:57OVAVXV257VDC5FZSSNPNMMUG","short_pith_number":"pith:57OVAVXV","schema_version":"1.0","canonical_sha256":"efdd5056f5d77f518ba5cca4d7b58ca19c9014ed094fceaba96b985ca011cb1b","source":{"kind":"arxiv","id":"2606.20582","version":1},"attestation_state":"computed","paper":{"title":"Communication-Semantic-Aware RDMA Loss Recovery for QP-scalable Hyperscale AI Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.NI","authors_text":"Fengyuan Ren, Jiakun Bao, Kun Zhu, Tong Zhang, Xiaoxiang Hua, Zhenjiang Dong","submitted_at":"2026-05-08T03:25:06Z","abstract_excerpt":"Current artificial intelligence (AI) infrastructures widely adopt Remote Direct Memory Access (RDMA) to support high-performance communication. Training trillion-parameter models involves frequent collective communication operations, such as All-Reduce and All-to-All, which generate intensive RDMA traffic. Existing RDMA deployments predominantly use the reliable connection (RC) model, where each process pair requires a dedicated queue pair (QP). This leads to poor scalability: since the RDMA-capable network interface card (RNIC) can cache only a few thousand QPs, excess entries trigger PCIe ro"},"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":"2606.20582","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NI","submitted_at":"2026-05-08T03:25:06Z","cross_cats_sorted":[],"title_canon_sha256":"b48280d6d927c039f7cf5f7461b77b449c0507e2ffe432002aa77cce70c48f5a","abstract_canon_sha256":"d215b97d3a577f02bb36f7bcaee12489462a1d633f5b25ca36e0809d8fed579c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T00:11:50.538851Z","signature_b64":"G4cDO5p8pf/nSnq/TLJX12AZ+MYMKs/o4l8ATFJbYtPqpsB/Tht/ShzKakXI0hINmUUcE98PCpCG2HqQoc3VCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"efdd5056f5d77f518ba5cca4d7b58ca19c9014ed094fceaba96b985ca011cb1b","last_reissued_at":"2026-06-23T00:11:50.538444Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T00:11:50.538444Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Communication-Semantic-Aware RDMA Loss Recovery for QP-scalable Hyperscale AI Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.NI","authors_text":"Fengyuan Ren, Jiakun Bao, Kun Zhu, Tong Zhang, Xiaoxiang Hua, Zhenjiang Dong","submitted_at":"2026-05-08T03:25:06Z","abstract_excerpt":"Current artificial intelligence (AI) infrastructures widely adopt Remote Direct Memory Access (RDMA) to support high-performance communication. Training trillion-parameter models involves frequent collective communication operations, such as All-Reduce and All-to-All, which generate intensive RDMA traffic. Existing RDMA deployments predominantly use the reliable connection (RC) model, where each process pair requires a dedicated queue pair (QP). This leads to poor scalability: since the RDMA-capable network interface card (RNIC) can cache only a few thousand QPs, excess entries trigger PCIe ro"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.20582","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/2606.20582/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":"2606.20582","created_at":"2026-06-23T00:11:50.538502+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.20582v1","created_at":"2026-06-23T00:11:50.538502+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.20582","created_at":"2026-06-23T00:11:50.538502+00:00"},{"alias_kind":"pith_short_12","alias_value":"57OVAVXV257V","created_at":"2026-06-23T00:11:50.538502+00:00"},{"alias_kind":"pith_short_16","alias_value":"57OVAVXV257VDC5F","created_at":"2026-06-23T00:11:50.538502+00:00"},{"alias_kind":"pith_short_8","alias_value":"57OVAVXV","created_at":"2026-06-23T00:11:50.538502+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/57OVAVXV257VDC5FZSSNPNMMUG","json":"https://pith.science/pith/57OVAVXV257VDC5FZSSNPNMMUG.json","graph_json":"https://pith.science/api/pith-number/57OVAVXV257VDC5FZSSNPNMMUG/graph.json","events_json":"https://pith.science/api/pith-number/57OVAVXV257VDC5FZSSNPNMMUG/events.json","paper":"https://pith.science/paper/57OVAVXV"},"agent_actions":{"view_html":"https://pith.science/pith/57OVAVXV257VDC5FZSSNPNMMUG","download_json":"https://pith.science/pith/57OVAVXV257VDC5FZSSNPNMMUG.json","view_paper":"https://pith.science/paper/57OVAVXV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.20582&json=true","fetch_graph":"https://pith.science/api/pith-number/57OVAVXV257VDC5FZSSNPNMMUG/graph.json","fetch_events":"https://pith.science/api/pith-number/57OVAVXV257VDC5FZSSNPNMMUG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/57OVAVXV257VDC5FZSSNPNMMUG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/57OVAVXV257VDC5FZSSNPNMMUG/action/storage_attestation","attest_author":"https://pith.science/pith/57OVAVXV257VDC5FZSSNPNMMUG/action/author_attestation","sign_citation":"https://pith.science/pith/57OVAVXV257VDC5FZSSNPNMMUG/action/citation_signature","submit_replication":"https://pith.science/pith/57OVAVXV257VDC5FZSSNPNMMUG/action/replication_record"}},"created_at":"2026-06-23T00:11:50.538502+00:00","updated_at":"2026-06-23T00:11:50.538502+00:00"}