{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:VIJRIHG4DBOWFBME552BZWH42N","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":"7d1317e00360c66671251d90076ec2582aac502b256e52c751b6db81155b5d7b","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.DC","submitted_at":"2025-03-12T18:08:01Z","title_canon_sha256":"174a11a2569e7cd39d0486b889bb8099a90d4e9d6ab1bcd38f0175426c54f224"},"schema_version":"1.0","source":{"id":"2503.09716","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.09716","created_at":"2026-07-05T10:30:06Z"},{"alias_kind":"arxiv_version","alias_value":"2503.09716v1","created_at":"2026-07-05T10:30:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.09716","created_at":"2026-07-05T10:30:06Z"},{"alias_kind":"pith_short_12","alias_value":"VIJRIHG4DBOW","created_at":"2026-07-05T10:30:06Z"},{"alias_kind":"pith_short_16","alias_value":"VIJRIHG4DBOWFBME","created_at":"2026-07-05T10:30:06Z"},{"alias_kind":"pith_short_8","alias_value":"VIJRIHG4","created_at":"2026-07-05T10:30:06Z"}],"graph_snapshots":[{"event_id":"sha256:bb68648d0232950190b2c574d0a8d9b414f31ce38c070d7dae7a64e19eaf7fe0","target":"graph","created_at":"2026-07-05T10:30:06Z","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/2503.09716/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"This paper presents MoE-Gen, a high-throughput MoE inference system optimized for single-GPU execution. Existing inference systems rely on model-based or continuous batching strategies, originally designed for interactive inference, which result in excessively small batches for MoE's key modules-attention and expert modules-leading to poor throughput. To address this, we introduce module-based batching, which accumulates tokens in host memory and dynamically launches large batches on GPUs to maximize utilization. Additionally, we optimize the choice of batch sizes for each module in an MoE to ","authors_text":"Adrian Jackson, Leyang Xue, Luo Mai, Tairan Xu, Zhan Lu","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.DC","submitted_at":"2025-03-12T18:08:01Z","title":"MoE-Gen: High-Throughput MoE Inference on a Single GPU with Module-Based Batching"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.09716","kind":"arxiv","version":1},"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":"sha256:923cb8ee4a2238458d2629c658721f017900ffce199b303d12d1a6b380a30cd8","target":"record","created_at":"2026-07-05T10:30:06Z","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":"7d1317e00360c66671251d90076ec2582aac502b256e52c751b6db81155b5d7b","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.DC","submitted_at":"2025-03-12T18:08:01Z","title_canon_sha256":"174a11a2569e7cd39d0486b889bb8099a90d4e9d6ab1bcd38f0175426c54f224"},"schema_version":"1.0","source":{"id":"2503.09716","kind":"arxiv","version":1}},"canonical_sha256":"aa13141cdc185d628584ef741cd8fcd36e9b0b571d2811585986aee5762d3e50","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"aa13141cdc185d628584ef741cd8fcd36e9b0b571d2811585986aee5762d3e50","first_computed_at":"2026-07-05T10:30:06.781169Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:30:06.781169Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"p+VQIvc8bP3Mley1AyBjznKtkLftYrTWlx79xPVhgN7KstGDBdbI0abY+zOcQsbmvEQ1YkcLhi6ebC1H+11pCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:30:06.781853Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.09716","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:923cb8ee4a2238458d2629c658721f017900ffce199b303d12d1a6b380a30cd8","sha256:bb68648d0232950190b2c574d0a8d9b414f31ce38c070d7dae7a64e19eaf7fe0"],"state_sha256":"8c9c11a7f6f259b2c1ceb54a7f18f3a9008ac284b98a763d835a9919cbe9bba3"}