{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:JA73RRPFS5T2CEDU3GG2VA46WI","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":"c5f2289a7b297ca5d66988aaad28ae92949230be64f4478c8bd248072767ee81","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2024-08-10T04:25:48Z","title_canon_sha256":"44d3937cf01a1572c907b96a0cee52b37bcddfdd3ae01c4f4fd3001c0d6d7338"},"schema_version":"1.0","source":{"id":"2408.05430","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2408.05430","created_at":"2026-07-05T08:54:20Z"},{"alias_kind":"arxiv_version","alias_value":"2408.05430v1","created_at":"2026-07-05T08:54:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.05430","created_at":"2026-07-05T08:54:20Z"},{"alias_kind":"pith_short_12","alias_value":"JA73RRPFS5T2","created_at":"2026-07-05T08:54:20Z"},{"alias_kind":"pith_short_16","alias_value":"JA73RRPFS5T2CEDU","created_at":"2026-07-05T08:54:20Z"},{"alias_kind":"pith_short_8","alias_value":"JA73RRPF","created_at":"2026-07-05T08:54:20Z"}],"graph_snapshots":[{"event_id":"sha256:c6dd97f4c6dba55bb5b0ede2576f7b438a18349d9f136674a697150c1916a1ba","target":"graph","created_at":"2026-07-05T08:54:20Z","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/2408.05430/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"In this paper, we present the practical problems and the lessons learned at short-video services from Kuaishou. In industry, a widely-used multi-task framework is the Mixture-of-Experts (MoE) paradigm, which always introduces some shared and specific experts for each task and then uses gate networks to measure related experts' contributions. Although the MoE achieves remarkable improvements, we still observe three anomalies that seriously affect model performances in our iteration: (1) Expert Collapse: We found that experts' output distributions are significantly different, and some experts ha","authors_text":"Guorui Zhou, Jiangxia Cao, Kun Gai, Xu Wang, Zhiyi Fu","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2024-08-10T04:25:48Z","title":"HoME: Hierarchy of Multi-Gate Experts for Multi-Task Learning at Kuaishou"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.05430","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":"sha256:0f796e14292745635dec05d7df62c5301e05c16f3fc5237f7c784864552442c9","target":"record","created_at":"2026-07-05T08:54:20Z","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":"c5f2289a7b297ca5d66988aaad28ae92949230be64f4478c8bd248072767ee81","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2024-08-10T04:25:48Z","title_canon_sha256":"44d3937cf01a1572c907b96a0cee52b37bcddfdd3ae01c4f4fd3001c0d6d7338"},"schema_version":"1.0","source":{"id":"2408.05430","kind":"arxiv","version":1}},"canonical_sha256":"483fb8c5e59767a11074d98daa839eb21fe2c15157a2cd2d75906b0141d0143e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"483fb8c5e59767a11074d98daa839eb21fe2c15157a2cd2d75906b0141d0143e","first_computed_at":"2026-07-05T08:54:20.166107Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:54:20.166107Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"T/5B8XLTN93DZjcLg/wIHJ7YKRetrZMe0OBFPSHaZjXKn8+TgTdMBDbLN+r8Hc6PxlA3nFlvwhXaEGKs0KqcBg==","signature_status":"signed_v1","signed_at":"2026-07-05T08:54:20.166503Z","signed_message":"canonical_sha256_bytes"},"source_id":"2408.05430","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0f796e14292745635dec05d7df62c5301e05c16f3fc5237f7c784864552442c9","sha256:c6dd97f4c6dba55bb5b0ede2576f7b438a18349d9f136674a697150c1916a1ba"],"state_sha256":"4c9df92a5cfa2a2835607a60361fe7fe3db5091035a29b13aa457fbb04b6a8f5"}