{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:WSOLXENYOG24XWYFBLA4HFBZWN","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"8439c402e56e198cd74ad0ba0b8c3e49c7953221b8dec8743a6d70d8e861bbd2","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2025-05-13T17:45:34Z","title_canon_sha256":"457a274a9b65d64181c65c3589b29991826b9b6d5e87b3ac1b019fe036491cb1"},"schema_version":"1.0","source":{"id":"2506.01969","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.01969","created_at":"2026-06-03T01:05:04Z"},{"alias_kind":"arxiv_version","alias_value":"2506.01969v3","created_at":"2026-06-03T01:05:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.01969","created_at":"2026-06-03T01:05:04Z"},{"alias_kind":"pith_short_12","alias_value":"WSOLXENYOG24","created_at":"2026-06-03T01:05:04Z"},{"alias_kind":"pith_short_16","alias_value":"WSOLXENYOG24XWYF","created_at":"2026-06-03T01:05:04Z"},{"alias_kind":"pith_short_8","alias_value":"WSOLXENY","created_at":"2026-06-03T01:05:04Z"}],"graph_snapshots":[{"event_id":"sha256:a6a53064cd1007c26451dd3ef43022c3a3f1ff6800a9f0f8fc2c636bb0a546fc","target":"graph","created_at":"2026-06-03T01:05:04Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2506.01969/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Efficient inference of Multi-Head Latent Attention (MLA) is challenged by deploying the DeepSeek-R1 671B model on a single Multi-GPU server. This paper introduces FlashMLA-ETAP, a novel framework that enhances MLA inference for the single-instance deployment scenario on NVIDIA H20 GPUs. We propose the Efficient Transpose Attention Pipeline (ETAP), which reconfigures attention computation through transposition to align the KV context length with the \\(M\\)-dimension in WGMMA operations, significantly reducing redundant computations. FlashMLA-ETAP achieves a 2.78x speedup over FlashMLA at 64K seq","authors_text":"Chang Kong, Pengcuo Dege, Qiuming Luo, Rui Mao","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2025-05-13T17:45:34Z","title":"FlashMLA-ETAP: Efficient Transpose Attention Pipeline for Accelerating MLA Inference on NVIDIA H20 GPUs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.01969","kind":"arxiv","version":3},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:b2cfe2ede27e5a0dc3d86d2e8e3e4e893e8b73419cf49f6da230c565a4929808","target":"record","created_at":"2026-06-03T01:05:04Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"8439c402e56e198cd74ad0ba0b8c3e49c7953221b8dec8743a6d70d8e861bbd2","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2025-05-13T17:45:34Z","title_canon_sha256":"457a274a9b65d64181c65c3589b29991826b9b6d5e87b3ac1b019fe036491cb1"},"schema_version":"1.0","source":{"id":"2506.01969","kind":"arxiv","version":3}},"canonical_sha256":"b49cbb91b871b5cbdb050ac1c39439b34a289074df3ed97592d57fa090aa3d96","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b49cbb91b871b5cbdb050ac1c39439b34a289074df3ed97592d57fa090aa3d96","first_computed_at":"2026-06-03T01:05:04.613339Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-03T01:05:04.613339Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"tp+gC28YSwhAH+r0B1P58HussTnSKLKkpoiJOLoaVCyDsnrdFiTqVk3UiX92xVonlhr35zdMnMWVhzJzabsQBA==","signature_status":"signed_v1","signed_at":"2026-06-03T01:05:04.613829Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.01969","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b2cfe2ede27e5a0dc3d86d2e8e3e4e893e8b73419cf49f6da230c565a4929808","sha256:a6a53064cd1007c26451dd3ef43022c3a3f1ff6800a9f0f8fc2c636bb0a546fc"],"state_sha256":"727e5b669a45a146b65b8c143a701ae16b0e8542f0c8287ef31420e8d7138fe8"}