{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:MO3T3DH2LSIM5XFZRFKEW25JE2","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":"4a7dede24ca46267e1480c7f0b6d57659ce66068c86487d3dd016d01526e528d","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-24T07:37:29Z","title_canon_sha256":"a1c2159ab191e515f656641ccfa8aed2d4158b624ea3e5c81a39bebb1248344b"},"schema_version":"1.0","source":{"id":"2502.16927","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.16927","created_at":"2026-07-05T10:26:14Z"},{"alias_kind":"arxiv_version","alias_value":"2502.16927v2","created_at":"2026-07-05T10:26:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.16927","created_at":"2026-07-05T10:26:14Z"},{"alias_kind":"pith_short_12","alias_value":"MO3T3DH2LSIM","created_at":"2026-07-05T10:26:14Z"},{"alias_kind":"pith_short_16","alias_value":"MO3T3DH2LSIM5XFZ","created_at":"2026-07-05T10:26:14Z"},{"alias_kind":"pith_short_8","alias_value":"MO3T3DH2","created_at":"2026-07-05T10:26:14Z"}],"graph_snapshots":[{"event_id":"sha256:efd71e4b38fbe05bcb266e369e99733a090ca39caf162d675733de4f6cda9175","target":"graph","created_at":"2026-07-05T10:26:14Z","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/2502.16927/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The Mixture-of-Experts (MoE) structure scales the Transformer-based large language models (LLMs) and improves their performance with only the sub-linear increase in computation resources. Recently, a fine-grained DeepSeekMoE structure is proposed, which can further improve the computing efficiency of MoE without performance degradation. However, the All-to-All communication introduced by MoE has become a bottleneck, especially for the fine-grained structure, which typically involves and activates more experts, hence contributing to heavier communication overhead.\n  In this paper, we propose a ","authors_text":"Cheng Li, Hongrui Zhan, Jiaan Zhu, Lin Zhang, Shengnan Wang, Youhui Bai, Zewen Jin, Zhenyu Ming","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-24T07:37:29Z","title":"BigMac: A Communication-Efficient Mixture-of-Experts Model Structure for Fast Training and Inference"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.16927","kind":"arxiv","version":2},"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":"sha256:064d729e145977a8cb5b909f1d68c31909ee6b5ff10f376dbaf6cb66f819a078","target":"record","created_at":"2026-07-05T10:26:14Z","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":"4a7dede24ca46267e1480c7f0b6d57659ce66068c86487d3dd016d01526e528d","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-24T07:37:29Z","title_canon_sha256":"a1c2159ab191e515f656641ccfa8aed2d4158b624ea3e5c81a39bebb1248344b"},"schema_version":"1.0","source":{"id":"2502.16927","kind":"arxiv","version":2}},"canonical_sha256":"63b73d8cfa5c90cedcb989544b6ba9269e1a9c372236f8f168dd137b7247c33e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"63b73d8cfa5c90cedcb989544b6ba9269e1a9c372236f8f168dd137b7247c33e","first_computed_at":"2026-07-05T10:26:14.532620Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:26:14.532620Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"HqJF/GBzdvJ//lXMqYpZ0SuOUTXcqM8kpHzlTSbeqUXuaNjtain7IO4SyZSmtUCYZ+hCDeaoCkduRpNd/xW/Dg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:26:14.533165Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.16927","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:064d729e145977a8cb5b909f1d68c31909ee6b5ff10f376dbaf6cb66f819a078","sha256:efd71e4b38fbe05bcb266e369e99733a090ca39caf162d675733de4f6cda9175"],"state_sha256":"7e402378030bc4b6db1ce91eab1b242c2cce8590e45b5803c57e23edf9619f43"}