{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:644WQAYFJNH7MOBID7FDMTIAJ7","short_pith_number":"pith:644WQAYF","schema_version":"1.0","canonical_sha256":"f7396803054b4ff638281fca364d004ffa4edfea3b060caa0591cc3ed706cd9f","source":{"kind":"arxiv","id":"2410.22134","version":3},"attestation_state":"computed","paper":{"title":"ProMoE: Fast MoE-based LLM Serving using Proactive Caching","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.DC","authors_text":"Haibo Chen, Rong chen, Xiaoniu Song, Zihang Zhong","submitted_at":"2024-10-29T15:31:27Z","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"},"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":"2410.22134","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2024-10-29T15:31:27Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"e0e9579aef02912eb8349c60e9ba9ccc94dc1d428d366c71d99711aa29662d12","abstract_canon_sha256":"dd9ac12dd3cf22b273ed4f0ea3e076714020c67380e6ad4b0a337555e38d6af2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:02:36.941305Z","signature_b64":"72V2dCQvpdtUJhshFORm+Kns+nhJp5O5OPl787mKXJnNqfTMhYJkMVl3FNySgSlkkB/DI6h41lDD0atAwkIeAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f7396803054b4ff638281fca364d004ffa4edfea3b060caa0591cc3ed706cd9f","last_reissued_at":"2026-07-05T12:02:36.940803Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:02:36.940803Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ProMoE: Fast MoE-based LLM Serving using Proactive Caching","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.DC","authors_text":"Haibo Chen, Rong chen, Xiaoniu Song, Zihang Zhong","submitted_at":"2024-10-29T15:31:27Z","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"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.22134","kind":"arxiv","version":3},"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/2410.22134/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":"2410.22134","created_at":"2026-07-05T12:02:36.940867+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.22134v3","created_at":"2026-07-05T12:02:36.940867+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.22134","created_at":"2026-07-05T12:02:36.940867+00:00"},{"alias_kind":"pith_short_12","alias_value":"644WQAYFJNH7","created_at":"2026-07-05T12:02:36.940867+00:00"},{"alias_kind":"pith_short_16","alias_value":"644WQAYFJNH7MOBI","created_at":"2026-07-05T12:02:36.940867+00:00"},{"alias_kind":"pith_short_8","alias_value":"644WQAYF","created_at":"2026-07-05T12:02:36.940867+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.10493","citing_title":"Achieving Cloud-Grade SLOs for Local Mixture-of-Experts Inference through CPU-GPU Hybrid Design","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29982","citing_title":"Beyond Uniform Experts: Cost-Aware Expert Execution for Efficient Multi-Device MoE Inference","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2601.21198","citing_title":"ZipMoE: Efficient On-Device MoE Serving via Lossless Compression and Cache-Affinity Scheduling","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2509.23638","citing_title":"LayerScope: Predictive Cross-Layer Scheduling for Efficient Multi-Batch MoE Inference on Legacy Servers","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05899","citing_title":"VisMMOE: Exploiting Visual-Expert Affinity for Efficient Visual-Language MoE Offloading","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20156","citing_title":"Temporally Extended Mixture-of-Experts Models","ref_index":36,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/644WQAYFJNH7MOBID7FDMTIAJ7","json":"https://pith.science/pith/644WQAYFJNH7MOBID7FDMTIAJ7.json","graph_json":"https://pith.science/api/pith-number/644WQAYFJNH7MOBID7FDMTIAJ7/graph.json","events_json":"https://pith.science/api/pith-number/644WQAYFJNH7MOBID7FDMTIAJ7/events.json","paper":"https://pith.science/paper/644WQAYF"},"agent_actions":{"view_html":"https://pith.science/pith/644WQAYFJNH7MOBID7FDMTIAJ7","download_json":"https://pith.science/pith/644WQAYFJNH7MOBID7FDMTIAJ7.json","view_paper":"https://pith.science/paper/644WQAYF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.22134&json=true","fetch_graph":"https://pith.science/api/pith-number/644WQAYFJNH7MOBID7FDMTIAJ7/graph.json","fetch_events":"https://pith.science/api/pith-number/644WQAYFJNH7MOBID7FDMTIAJ7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/644WQAYFJNH7MOBID7FDMTIAJ7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/644WQAYFJNH7MOBID7FDMTIAJ7/action/storage_attestation","attest_author":"https://pith.science/pith/644WQAYFJNH7MOBID7FDMTIAJ7/action/author_attestation","sign_citation":"https://pith.science/pith/644WQAYFJNH7MOBID7FDMTIAJ7/action/citation_signature","submit_replication":"https://pith.science/pith/644WQAYFJNH7MOBID7FDMTIAJ7/action/replication_record"}},"created_at":"2026-07-05T12:02:36.940867+00:00","updated_at":"2026-07-05T12:02:36.940867+00:00"}