{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:644WQAYFJNH7MOBID7FDMTIAJ7","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":"dd9ac12dd3cf22b273ed4f0ea3e076714020c67380e6ad4b0a337555e38d6af2","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2024-10-29T15:31:27Z","title_canon_sha256":"e0e9579aef02912eb8349c60e9ba9ccc94dc1d428d366c71d99711aa29662d12"},"schema_version":"1.0","source":{"id":"2410.22134","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.22134","created_at":"2026-07-05T12:02:36Z"},{"alias_kind":"arxiv_version","alias_value":"2410.22134v3","created_at":"2026-07-05T12:02:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.22134","created_at":"2026-07-05T12:02:36Z"},{"alias_kind":"pith_short_12","alias_value":"644WQAYFJNH7","created_at":"2026-07-05T12:02:36Z"},{"alias_kind":"pith_short_16","alias_value":"644WQAYFJNH7MOBI","created_at":"2026-07-05T12:02:36Z"},{"alias_kind":"pith_short_8","alias_value":"644WQAYF","created_at":"2026-07-05T12:02:36Z"}],"graph_snapshots":[{"event_id":"sha256:3913f77e994d3c9031ec5d8feb9be64e986dec9d547fe2127a56566c122d537b","target":"graph","created_at":"2026-07-05T12:02:36Z","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/2410.22134/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The promising applications of large language models are often limited by the constrained GPU memory capacity available on edge devices. Mixture-of-Experts (MoE) models help address this issue by activating only a subset of the model's parameters during computation. This approach allows the unused parameters to be offloaded to host memory, thereby reducing the overall GPU memory demand. However, existing cache-based offloading solutions handle cache misses reactively, which significantly impacts system performance. In this paper, we introduce ProMoE, a novel proactive caching system that utiliz","authors_text":"Haibo Chen, Rong chen, Xiaoniu Song, Zihang Zhong","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2024-10-29T15:31:27Z","title":"ProMoE: Fast MoE-based LLM Serving using Proactive Caching"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.22134","kind":"arxiv","version":3},"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":"sha256:60a51eff7797efae3e8cb1eed6baa05c36983ea5070a6d179933ec181b93e531","target":"record","created_at":"2026-07-05T12:02:36Z","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":"dd9ac12dd3cf22b273ed4f0ea3e076714020c67380e6ad4b0a337555e38d6af2","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2024-10-29T15:31:27Z","title_canon_sha256":"e0e9579aef02912eb8349c60e9ba9ccc94dc1d428d366c71d99711aa29662d12"},"schema_version":"1.0","source":{"id":"2410.22134","kind":"arxiv","version":3}},"canonical_sha256":"f7396803054b4ff638281fca364d004ffa4edfea3b060caa0591cc3ed706cd9f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f7396803054b4ff638281fca364d004ffa4edfea3b060caa0591cc3ed706cd9f","first_computed_at":"2026-07-05T12:02:36.940803Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:02:36.940803Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"72V2dCQvpdtUJhshFORm+Kns+nhJp5O5OPl787mKXJnNqfTMhYJkMVl3FNySgSlkkB/DI6h41lDD0atAwkIeAw==","signature_status":"signed_v1","signed_at":"2026-07-05T12:02:36.941305Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.22134","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:60a51eff7797efae3e8cb1eed6baa05c36983ea5070a6d179933ec181b93e531","sha256:3913f77e994d3c9031ec5d8feb9be64e986dec9d547fe2127a56566c122d537b"],"state_sha256":"30ee698b42fdf34713cc4012c62f5dc37d7db76534e1186b133230e38b433b0f"}