{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:56ZO75EUQK3I5TWLLUO5PCDX34","short_pith_number":"pith:56ZO75EU","schema_version":"1.0","canonical_sha256":"efb2eff49482b68ececb5d1dd78877df0f90879c87ad0c9225546fcfd1fbb1bd","source":{"kind":"arxiv","id":"2410.17043","version":1},"attestation_state":"computed","paper":{"title":"Optimizing Mixture-of-Experts Inference Time Combining Model Deployment and Communication Scheduling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NI"],"primary_cat":"cs.LG","authors_text":"Jialong Li, Lakshay Rastogi, Rui Pan, Shreyansh Tripathi, Yiming Lei, Yiting Xia","submitted_at":"2024-10-22T14:19:29Z","abstract_excerpt":"As machine learning models scale in size and complexity, their computational requirements become a significant barrier. Mixture-of-Experts (MoE) models alleviate this issue by selectively activating relevant experts. Despite this, MoE models are hindered by high communication overhead from all-to-all operations, low GPU utilization due to the synchronous communication constraint, and complications from heterogeneous GPU environments.\n  This paper presents Aurora, which optimizes both model deployment and all-to-all communication scheduling to address these challenges in MoE inference. Aurora a"},"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":"2410.17043","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-10-22T14:19:29Z","cross_cats_sorted":["cs.NI"],"title_canon_sha256":"da253c3362cca667a1e6d09317acfe6187fa8ad081ab8fa4cdf16959ccfe0b64","abstract_canon_sha256":"042bceb2c0e8c0642c59c59c5e52ba4fcb7888bdf020419ce87e3c070d80e3bc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:24:11.175228Z","signature_b64":"s4eVD2qL/I3pJIwykmYVnxjLL3kiyyEdrcURg6G26Ip8NbpEOc0e86EWnBsRjyFjXyi1ZHbjJymekL20DwXbAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"efb2eff49482b68ececb5d1dd78877df0f90879c87ad0c9225546fcfd1fbb1bd","last_reissued_at":"2026-07-05T09:24:11.174711Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:24:11.174711Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Optimizing Mixture-of-Experts Inference Time Combining Model Deployment and Communication Scheduling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NI"],"primary_cat":"cs.LG","authors_text":"Jialong Li, Lakshay Rastogi, Rui Pan, Shreyansh Tripathi, Yiming Lei, Yiting Xia","submitted_at":"2024-10-22T14:19:29Z","abstract_excerpt":"As machine learning models scale in size and complexity, their computational requirements become a significant barrier. Mixture-of-Experts (MoE) models alleviate this issue by selectively activating relevant experts. Despite this, MoE models are hindered by high communication overhead from all-to-all operations, low GPU utilization due to the synchronous communication constraint, and complications from heterogeneous GPU environments.\n  This paper presents Aurora, which optimizes both model deployment and all-to-all communication scheduling to address these challenges in MoE inference. Aurora a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.17043","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/2410.17043/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":"2410.17043","created_at":"2026-07-05T09:24:11.174781+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.17043v1","created_at":"2026-07-05T09:24:11.174781+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.17043","created_at":"2026-07-05T09:24:11.174781+00:00"},{"alias_kind":"pith_short_12","alias_value":"56ZO75EUQK3I","created_at":"2026-07-05T09:24:11.174781+00:00"},{"alias_kind":"pith_short_16","alias_value":"56ZO75EUQK3I5TWL","created_at":"2026-07-05T09:24:11.174781+00:00"},{"alias_kind":"pith_short_8","alias_value":"56ZO75EU","created_at":"2026-07-05T09:24:11.174781+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.00515","citing_title":"SpaceMoE: Realizing Distributed Mixture-of-Experts Inference over Space Networks","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2510.05497","citing_title":"Patterns behind Chaos: Forecasting Data Movement for Efficient Large-Scale MoE LLM Inference","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00515","citing_title":"SpaceMoE: Realizing Distributed Mixture-of-Experts Inference over Space Networks","ref_index":15,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/56ZO75EUQK3I5TWLLUO5PCDX34","json":"https://pith.science/pith/56ZO75EUQK3I5TWLLUO5PCDX34.json","graph_json":"https://pith.science/api/pith-number/56ZO75EUQK3I5TWLLUO5PCDX34/graph.json","events_json":"https://pith.science/api/pith-number/56ZO75EUQK3I5TWLLUO5PCDX34/events.json","paper":"https://pith.science/paper/56ZO75EU"},"agent_actions":{"view_html":"https://pith.science/pith/56ZO75EUQK3I5TWLLUO5PCDX34","download_json":"https://pith.science/pith/56ZO75EUQK3I5TWLLUO5PCDX34.json","view_paper":"https://pith.science/paper/56ZO75EU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.17043&json=true","fetch_graph":"https://pith.science/api/pith-number/56ZO75EUQK3I5TWLLUO5PCDX34/graph.json","fetch_events":"https://pith.science/api/pith-number/56ZO75EUQK3I5TWLLUO5PCDX34/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/56ZO75EUQK3I5TWLLUO5PCDX34/action/timestamp_anchor","attest_storage":"https://pith.science/pith/56ZO75EUQK3I5TWLLUO5PCDX34/action/storage_attestation","attest_author":"https://pith.science/pith/56ZO75EUQK3I5TWLLUO5PCDX34/action/author_attestation","sign_citation":"https://pith.science/pith/56ZO75EUQK3I5TWLLUO5PCDX34/action/citation_signature","submit_replication":"https://pith.science/pith/56ZO75EUQK3I5TWLLUO5PCDX34/action/replication_record"}},"created_at":"2026-07-05T09:24:11.174781+00:00","updated_at":"2026-07-05T09:24:11.174781+00:00"}