{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:SGRHL77BNAP25FGDUODECICWAT","short_pith_number":"pith:SGRHL77B","schema_version":"1.0","canonical_sha256":"91a275ffe1681fae94c3a38641205604f0c2f8207ac4823a618be80750689201","source":{"kind":"arxiv","id":"2605.02960","version":2},"attestation_state":"computed","paper":{"title":"MoE-Prefill: Zero Redundancy Overheads in MoE Prefill Serving","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"MoE prefill serving eliminates redundant overheads by asynchronously gathering expert weights during compute-bound phases.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Aurick Qiao, Juncheng Yang, Karthik Ganesan, Olatunji Ruwase, Samyam Rajbhandari, Yue Cheng, Yuxiong He, Zhaoyuan Su","submitted_at":"2026-05-03T03:10:24Z","abstract_excerpt":"Production LLM workloads increasingly serve discriminative tasks, such as classification, recommendation, and verification, whose answers are read from the logits of a single prefill pass with no autoregressive decoding. Serving these prefill-only workloads on mixture-of-experts (MoE) models is bottlenecked not by compute but by the distributed execution required to fit the model: existing parallel strategies (tensor, expert, and pipeline parallelism) trade memory pressure for redundant computation, communication, and synchronization, severely degrading MoE prefill serving efficiency. We obser"},"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":true,"formal_links_present":false},"canonical_record":{"source":{"id":"2605.02960","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-03T03:10:24Z","cross_cats_sorted":[],"title_canon_sha256":"8f2d740a74c55c27cdcf58160f8ba9005e393ea49b602d7e6f62e60564e6bf54","abstract_canon_sha256":"bf1f9ca812600cc95e0be27a7a4414028f4b550c704c5c74d72538c538b50d81"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:00:40.478275Z","signature_b64":"YQEMyM8sArO+sU2UNbVzYk5GDQaVllId1AZkhparJvI3X0J1LF9/7bicYaLQsx0KX0wWZtyu2wzWNwjElGBHAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"91a275ffe1681fae94c3a38641205604f0c2f8207ac4823a618be80750689201","last_reissued_at":"2026-05-20T00:00:40.477498Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:00:40.477498Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MoE-Prefill: Zero Redundancy Overheads in MoE Prefill Serving","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"MoE prefill serving eliminates redundant overheads by asynchronously gathering expert weights during compute-bound phases.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Aurick Qiao, Juncheng Yang, Karthik Ganesan, Olatunji Ruwase, Samyam Rajbhandari, Yue Cheng, Yuxiong He, Zhaoyuan Su","submitted_at":"2026-05-03T03:10:24Z","abstract_excerpt":"Production LLM workloads increasingly serve discriminative tasks, such as classification, recommendation, and verification, whose answers are read from the logits of a single prefill pass with no autoregressive decoding. Serving these prefill-only workloads on mixture-of-experts (MoE) models is bottlenecked not by compute but by the distributed execution required to fit the model: existing parallel strategies (tensor, expert, and pipeline parallelism) trade memory pressure for redundant computation, communication, and synchronization, severely degrading MoE prefill serving efficiency. We obser"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"On Qwen3-235B-A22B across four hardware/precision configurations, MoE-Prefill delivers 1.35-1.37x throughput over the strongest distributed baseline on real-world workloads and up to 1.59x on long-context synthetic workloads, sustaining 29.8-36.2% per-GPU model FLOPs utilization.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The long, compute-bound forward passes of large-batch prefill open a per-layer window wide enough to stream expert weights in the background, replacing per-layer activation AllToAll with asynchronous weight AllGather fully overlapped with computation without new bottlenecks or accuracy loss.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"MoE-Prefill achieves 1.35-1.59x higher throughput for prefill-only MoE serving by using asynchronous expert parallelism to overlap weight AllGather with computation and prefix-aware routing with true-FLOPs tracking.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"MoE prefill serving eliminates redundant overheads by asynchronously gathering expert weights during compute-bound phases.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"74b8371a6a6832046d254f75f767c3ebd917729fcbae67b252d77215f264f953"},"source":{"id":"2605.02960","kind":"arxiv","version":2},"verdict":{"id":"83754aca-5224-4f93-b30f-082eafe82638","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T17:38:54.747167Z","strongest_claim":"On Qwen3-235B-A22B across four hardware/precision configurations, MoE-Prefill delivers 1.35-1.37x throughput over the strongest distributed baseline on real-world workloads and up to 1.59x on long-context synthetic workloads, sustaining 29.8-36.2% per-GPU model FLOPs utilization.","one_line_summary":"MoE-Prefill achieves 1.35-1.59x higher throughput for prefill-only MoE serving by using asynchronous expert parallelism to overlap weight AllGather with computation and prefix-aware routing with true-FLOPs tracking.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The long, compute-bound forward passes of large-batch prefill open a per-layer window wide enough to stream expert weights in the background, replacing per-layer activation AllToAll with asynchronous weight AllGather fully overlapped with computation without new bottlenecks or accuracy loss.","pith_extraction_headline":"MoE prefill serving eliminates redundant overheads by asynchronously gathering expert weights during compute-bound phases."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.02960/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_compliance","ran_at":"2026-05-19T17:01:31.756960Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"a1878eac2e9beaf64465e0bee886c01093de3f9ce883ca40f498e2da03b942f6"},"references":{"count":75,"sample":[{"doi":"","year":2025,"title":"gpt-oss-120b & gpt-oss-20b Model Card","work_id":"178c1f7e-4f19-4392-a45d-45a6dfa88ead","ref_index":1,"cited_arxiv_id":"2508.10925","is_internal_anchor":true},{"doi":"","year":2023,"title":"SARATHI: Efficient LLM Inference by Piggybacking Decodes with Chunked Prefills","work_id":"3dbdd757-ca01-436f-acfd-12ffcd6f64c6","ref_index":2,"cited_arxiv_id":"2308.16369","is_internal_anchor":true},{"doi":"","year":2022,"title":"Deepspeed-inference: enabling efficient in- ference of transformer models at unprecedented scale","work_id":"461bf7aa-e1b9-4309-a540-e2905d823c31","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"A survey on mixture of experts in large language models.IEEE Transactions on Knowledge and Data Engineering, 2025","work_id":"c76fa220-c8af-40e4-a69f-ee8ff33e469b","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Moe-lightning: High-throughput moe inference on memory-constrained gpus","work_id":"c7c2193d-2846-4a86-8642-116ade28549b","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":75,"snapshot_sha256":"e667420ecd0ad681b43542be402886428e8f63d7873ba26aa2adae8ae08a04b0","internal_anchors":16},"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":"2605.02960","created_at":"2026-05-20T00:00:40.477634+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.02960v2","created_at":"2026-05-20T00:00:40.477634+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.02960","created_at":"2026-05-20T00:00:40.477634+00:00"},{"alias_kind":"pith_short_12","alias_value":"SGRHL77BNAP2","created_at":"2026-05-20T00:00:40.477634+00:00"},{"alias_kind":"pith_short_16","alias_value":"SGRHL77BNAP25FGD","created_at":"2026-05-20T00:00:40.477634+00:00"},{"alias_kind":"pith_short_8","alias_value":"SGRHL77B","created_at":"2026-05-20T00:00:40.477634+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/SGRHL77BNAP25FGDUODECICWAT","json":"https://pith.science/pith/SGRHL77BNAP25FGDUODECICWAT.json","graph_json":"https://pith.science/api/pith-number/SGRHL77BNAP25FGDUODECICWAT/graph.json","events_json":"https://pith.science/api/pith-number/SGRHL77BNAP25FGDUODECICWAT/events.json","paper":"https://pith.science/paper/SGRHL77B"},"agent_actions":{"view_html":"https://pith.science/pith/SGRHL77BNAP25FGDUODECICWAT","download_json":"https://pith.science/pith/SGRHL77BNAP25FGDUODECICWAT.json","view_paper":"https://pith.science/paper/SGRHL77B","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.02960&json=true","fetch_graph":"https://pith.science/api/pith-number/SGRHL77BNAP25FGDUODECICWAT/graph.json","fetch_events":"https://pith.science/api/pith-number/SGRHL77BNAP25FGDUODECICWAT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SGRHL77BNAP25FGDUODECICWAT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SGRHL77BNAP25FGDUODECICWAT/action/storage_attestation","attest_author":"https://pith.science/pith/SGRHL77BNAP25FGDUODECICWAT/action/author_attestation","sign_citation":"https://pith.science/pith/SGRHL77BNAP25FGDUODECICWAT/action/citation_signature","submit_replication":"https://pith.science/pith/SGRHL77BNAP25FGDUODECICWAT/action/replication_record"}},"created_at":"2026-05-20T00:00:40.477634+00:00","updated_at":"2026-05-20T00:00:40.477634+00:00"}