{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:CZBAALA2CBMFA3DAHKSN26PYVD","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":"c4766544cad8f6e0ffb2279a3cc6c123784f9223fd0f3c55e6898eba4200e32f","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2023-08-29T05:25:21Z","title_canon_sha256":"0351f87f1934bcb3b12863e4fd246ba9ccb6e4a484f54b45251472e32160bae9"},"schema_version":"1.0","source":{"id":"2308.15030","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2308.15030","created_at":"2026-07-05T08:24:37Z"},{"alias_kind":"arxiv_version","alias_value":"2308.15030v4","created_at":"2026-07-05T08:24:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2308.15030","created_at":"2026-07-05T08:24:37Z"},{"alias_kind":"pith_short_12","alias_value":"CZBAALA2CBMF","created_at":"2026-07-05T08:24:37Z"},{"alias_kind":"pith_short_16","alias_value":"CZBAALA2CBMFA3DA","created_at":"2026-07-05T08:24:37Z"},{"alias_kind":"pith_short_8","alias_value":"CZBAALA2","created_at":"2026-07-05T08:24:37Z"}],"graph_snapshots":[{"event_id":"sha256:7a693b231f3c33ee12fd777d4b4a1b60ff6a6fee96a8f268f3e6007cc60bc01a","target":"graph","created_at":"2026-07-05T08:24:37Z","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/2308.15030/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Mixture of experts (MoE) is a popular technique to improve capacity of Large Language Models (LLMs) with conditionally-activated parallel experts. However, serving MoE models on memory-constrained devices is challenging due to the large parameter size. Typical solutions such as memory swapping or expert pruning may lead to significantly higher latency or severe accuracy loss. In this paper, we introduce SwapMoE, a framework for efficient serving of MoE-based large language models with tunable memory budgets. The main idea of SwapMoE is to keep a small dynamic set of important experts, namely V","authors_text":"Linghe Kong, Qingtian Feng, Rui Kong, Weijun Wang, Xiaozhou Ye, Ye Ouyang, Yuanchun Li, Yunxin Liu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2023-08-29T05:25:21Z","title":"SwapMoE: Serving Off-the-shelf MoE-based Large Language Models with Tunable Memory Budget"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2308.15030","kind":"arxiv","version":4},"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":"sha256:0677f6015bc81531841a8621c083cd3b919356b1f50715047b1564a21ce736e4","target":"record","created_at":"2026-07-05T08:24:37Z","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":"c4766544cad8f6e0ffb2279a3cc6c123784f9223fd0f3c55e6898eba4200e32f","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2023-08-29T05:25:21Z","title_canon_sha256":"0351f87f1934bcb3b12863e4fd246ba9ccb6e4a484f54b45251472e32160bae9"},"schema_version":"1.0","source":{"id":"2308.15030","kind":"arxiv","version":4}},"canonical_sha256":"1642002c1a1058506c603aa4dd79f8a8dc51be5aaba6f0e95f348c199df5754c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1642002c1a1058506c603aa4dd79f8a8dc51be5aaba6f0e95f348c199df5754c","first_computed_at":"2026-07-05T08:24:37.456588Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:24:37.456588Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Jtumei95LNKUE63dduyvQbCAB4oBAzc5YNfIxO2t+K6PlJnTMaSk/dquy1c4VZwBgYR0XHRoVIecXIbnqMK1AA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:24:37.457094Z","signed_message":"canonical_sha256_bytes"},"source_id":"2308.15030","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0677f6015bc81531841a8621c083cd3b919356b1f50715047b1564a21ce736e4","sha256:7a693b231f3c33ee12fd777d4b4a1b60ff6a6fee96a8f268f3e6007cc60bc01a"],"state_sha256":"ffcc1a04a6fe8a1072b1bb07f8e1e8251e887f62a3dacffd1ce7c4ebf05d3da0"}