{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:PLBFLFHSGKG3CYUZKSWKIEX35Y","short_pith_number":"pith:PLBFLFHS","schema_version":"1.0","canonical_sha256":"7ac25594f2328db1629954aca412fbee358cef718b6df448187df55ad7a057be","source":{"kind":"arxiv","id":"2606.01007","version":1},"attestation_state":"computed","paper":{"title":"Beyond Task-Agnostic: Task-Aware Grouping for Communication-Efficient Multi-Task MoE Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Aoxue Liu, Dan Zhao, Qing Li, Yong Jiang, Zhanjie Ding, Zhiyao Xu","submitted_at":"2026-05-31T04:51:09Z","abstract_excerpt":"Sparsely activated Mixture-of-Experts (MoE) models scale capacity via conditional computation, but distributed inference suffers from cross-GPU expert communication and routing-induced load imbalance. Existing placement methods reduce this cost by co-locating frequently co-activated experts; however, they derive a single deployment plan from globally aggregated routing traces, thereby averaging away the heterogeneous, task-specific co-activation patterns that actually drive communication in multi-task serving. We observe that expert co-activation is strongly task-conditioned: pairs tightly cou"},"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":"2606.01007","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-31T04:51:09Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2efe9a73763b13b3fcb07ed26a73e5b562891f9d1fecb5c45fcf5c8c006ebcca","abstract_canon_sha256":"06cccc64b8397daee17662790b6e63d1f8ff1f1a4696a33bd0e7ca49c2c733d7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T01:04:18.171031Z","signature_b64":"hyBwFxGu1k/GKwGUsB9rvguLeDvOt1UfPtI8CQcTZaypG6Y6Tmqn/N1VUx1vOmADiEc4N7EhJM5nmrpYmhEGAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7ac25594f2328db1629954aca412fbee358cef718b6df448187df55ad7a057be","last_reissued_at":"2026-06-02T01:04:18.170617Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T01:04:18.170617Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond Task-Agnostic: Task-Aware Grouping for Communication-Efficient Multi-Task MoE Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Aoxue Liu, Dan Zhao, Qing Li, Yong Jiang, Zhanjie Ding, Zhiyao Xu","submitted_at":"2026-05-31T04:51:09Z","abstract_excerpt":"Sparsely activated Mixture-of-Experts (MoE) models scale capacity via conditional computation, but distributed inference suffers from cross-GPU expert communication and routing-induced load imbalance. Existing placement methods reduce this cost by co-locating frequently co-activated experts; however, they derive a single deployment plan from globally aggregated routing traces, thereby averaging away the heterogeneous, task-specific co-activation patterns that actually drive communication in multi-task serving. We observe that expert co-activation is strongly task-conditioned: pairs tightly cou"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.01007","kind":"arxiv","version":1},"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/2606.01007/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":"2606.01007","created_at":"2026-06-02T01:04:18.170677+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.01007v1","created_at":"2026-06-02T01:04:18.170677+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.01007","created_at":"2026-06-02T01:04:18.170677+00:00"},{"alias_kind":"pith_short_12","alias_value":"PLBFLFHSGKG3","created_at":"2026-06-02T01:04:18.170677+00:00"},{"alias_kind":"pith_short_16","alias_value":"PLBFLFHSGKG3CYUZ","created_at":"2026-06-02T01:04:18.170677+00:00"},{"alias_kind":"pith_short_8","alias_value":"PLBFLFHS","created_at":"2026-06-02T01:04:18.170677+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PLBFLFHSGKG3CYUZKSWKIEX35Y","json":"https://pith.science/pith/PLBFLFHSGKG3CYUZKSWKIEX35Y.json","graph_json":"https://pith.science/api/pith-number/PLBFLFHSGKG3CYUZKSWKIEX35Y/graph.json","events_json":"https://pith.science/api/pith-number/PLBFLFHSGKG3CYUZKSWKIEX35Y/events.json","paper":"https://pith.science/paper/PLBFLFHS"},"agent_actions":{"view_html":"https://pith.science/pith/PLBFLFHSGKG3CYUZKSWKIEX35Y","download_json":"https://pith.science/pith/PLBFLFHSGKG3CYUZKSWKIEX35Y.json","view_paper":"https://pith.science/paper/PLBFLFHS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.01007&json=true","fetch_graph":"https://pith.science/api/pith-number/PLBFLFHSGKG3CYUZKSWKIEX35Y/graph.json","fetch_events":"https://pith.science/api/pith-number/PLBFLFHSGKG3CYUZKSWKIEX35Y/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PLBFLFHSGKG3CYUZKSWKIEX35Y/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PLBFLFHSGKG3CYUZKSWKIEX35Y/action/storage_attestation","attest_author":"https://pith.science/pith/PLBFLFHSGKG3CYUZKSWKIEX35Y/action/author_attestation","sign_citation":"https://pith.science/pith/PLBFLFHSGKG3CYUZKSWKIEX35Y/action/citation_signature","submit_replication":"https://pith.science/pith/PLBFLFHSGKG3CYUZKSWKIEX35Y/action/replication_record"}},"created_at":"2026-06-02T01:04:18.170677+00:00","updated_at":"2026-06-02T01:04:18.170677+00:00"}