{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:MQYT6EG66VGMPJVLQOY2V5OMIO","short_pith_number":"pith:MQYT6EG6","schema_version":"1.0","canonical_sha256":"64313f10def54cc7a6ab83b1aaf5cc43b017eb50c4f75ea2f054d40ed9f9c027","source":{"kind":"arxiv","id":"2607.04164","version":1},"attestation_state":"computed","paper":{"title":"BrownoutMoE: Structure-Aware Expert Grouping for Efficient and Accurate LLM Web-based Services","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Chengzhong Xu, Kejiang Ye, Minxian Xu, Yi Ding, Zhengxin Fang","submitted_at":"2026-07-05T08:08:26Z","abstract_excerpt":"Mixture-of-Experts (MoE) large language models (LLMs) are increasingly deployed in Web-facing services, where inference must be both accurate and responsive under bursty demand. Although MoE models improve parameter efficiency through sparse expert activation, efficient MoE inference remains challenging in practice. A major reason is the highly imbalanced expert access pattern during inference: a few hot experts process most routed tokens, while many cold experts are rarely activated, leaving GPU parallelism underutilized. Existing systems mainly optimize runtime execution, such as scheduling,"},"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":"2607.04164","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2026-07-05T08:08:26Z","cross_cats_sorted":[],"title_canon_sha256":"b38fd2e325568f603df90a0bbb37e2b671b8f0a68e6067b83d3869e6cca17cb4","abstract_canon_sha256":"2adb05fe1c5dd59013be6737d84c154a0f32a2620875a02d3c4562384b0a97e5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-07T02:19:00.531116Z","signature_b64":"8a/oIH+UpR5qNRYeZZSyaH/Gh1+uGVc1qTWbfM9rZf21h5ElEijd2uyU5Nu8jw4DzjJM42EN2i1sA/5dM4gSAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"64313f10def54cc7a6ab83b1aaf5cc43b017eb50c4f75ea2f054d40ed9f9c027","last_reissued_at":"2026-07-07T02:19:00.530397Z","signature_status":"signed_v1","first_computed_at":"2026-07-07T02:19:00.530397Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BrownoutMoE: Structure-Aware Expert Grouping for Efficient and Accurate LLM Web-based Services","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Chengzhong Xu, Kejiang Ye, Minxian Xu, Yi Ding, Zhengxin Fang","submitted_at":"2026-07-05T08:08:26Z","abstract_excerpt":"Mixture-of-Experts (MoE) large language models (LLMs) are increasingly deployed in Web-facing services, where inference must be both accurate and responsive under bursty demand. Although MoE models improve parameter efficiency through sparse expert activation, efficient MoE inference remains challenging in practice. A major reason is the highly imbalanced expert access pattern during inference: a few hot experts process most routed tokens, while many cold experts are rarely activated, leaving GPU parallelism underutilized. Existing systems mainly optimize runtime execution, such as scheduling,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.04164","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/2607.04164/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":"2607.04164","created_at":"2026-07-07T02:19:00.530508+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.04164v1","created_at":"2026-07-07T02:19:00.530508+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.04164","created_at":"2026-07-07T02:19:00.530508+00:00"},{"alias_kind":"pith_short_12","alias_value":"MQYT6EG66VGM","created_at":"2026-07-07T02:19:00.530508+00:00"},{"alias_kind":"pith_short_16","alias_value":"MQYT6EG66VGMPJVL","created_at":"2026-07-07T02:19:00.530508+00:00"},{"alias_kind":"pith_short_8","alias_value":"MQYT6EG6","created_at":"2026-07-07T02:19:00.530508+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/MQYT6EG66VGMPJVLQOY2V5OMIO","json":"https://pith.science/pith/MQYT6EG66VGMPJVLQOY2V5OMIO.json","graph_json":"https://pith.science/api/pith-number/MQYT6EG66VGMPJVLQOY2V5OMIO/graph.json","events_json":"https://pith.science/api/pith-number/MQYT6EG66VGMPJVLQOY2V5OMIO/events.json","paper":"https://pith.science/paper/MQYT6EG6"},"agent_actions":{"view_html":"https://pith.science/pith/MQYT6EG66VGMPJVLQOY2V5OMIO","download_json":"https://pith.science/pith/MQYT6EG66VGMPJVLQOY2V5OMIO.json","view_paper":"https://pith.science/paper/MQYT6EG6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.04164&json=true","fetch_graph":"https://pith.science/api/pith-number/MQYT6EG66VGMPJVLQOY2V5OMIO/graph.json","fetch_events":"https://pith.science/api/pith-number/MQYT6EG66VGMPJVLQOY2V5OMIO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MQYT6EG66VGMPJVLQOY2V5OMIO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MQYT6EG66VGMPJVLQOY2V5OMIO/action/storage_attestation","attest_author":"https://pith.science/pith/MQYT6EG66VGMPJVLQOY2V5OMIO/action/author_attestation","sign_citation":"https://pith.science/pith/MQYT6EG66VGMPJVLQOY2V5OMIO/action/citation_signature","submit_replication":"https://pith.science/pith/MQYT6EG66VGMPJVLQOY2V5OMIO/action/replication_record"}},"created_at":"2026-07-07T02:19:00.530508+00:00","updated_at":"2026-07-07T02:19:00.530508+00:00"}