{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:SZHO6TLYHUTTTKM4TM2XSZKGPT","short_pith_number":"pith:SZHO6TLY","schema_version":"1.0","canonical_sha256":"964eef4d783d2739a99c9b357965467cc56789c6ac99c87f31ea4fb80152696e","source":{"kind":"arxiv","id":"2608.12146","version":1},"attestation_state":"computed","paper":{"title":"RoutePack: Expert Placement and Attention-Aware Data Packing for MoE Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Gen Li, Xiaowei Zhu, Xudong Han, Yibo Shen, Zhenxuan Pan","submitted_at":"2026-08-12T15:07:15Z","abstract_excerpt":"Training Mixture-of-Experts (MoE) models for reinforcement learning (RL) couples two load-balancing problems: sequence composition determines dense attention work in each data-parallel microbatch, while token routing determines sparse expert work on expert-parallel ranks. Optimizing either alone can shift the bottleneck to the other. In MoE RL, rollout-time routing replay exposes every sample's sequence length and layer-wise expert demand before its training step. We present RoutePack, a hierarchical planner that coordinates state-consistent, layer-wise expert rerouting with joint attention- a"},"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":"2608.12146","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2026-08-12T15:07:15Z","cross_cats_sorted":[],"title_canon_sha256":"40f71bb5d57714ac275efc39ac3c391d552ee9a278d60c53ded108239de1c7a3","abstract_canon_sha256":"b70a9ec296f9c73cd116bb1a9d1752cbb0954c8cb6fb65a35f202e7cde6112b6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-13T01:30:13.566559Z","signature_b64":"c6GAqqNSoEONkM3kzErDlRWLFjlv3OixnpqCqs/mkamRGRgcvDnhEIoY9HC1wpCe7hUYp4Myll2naTsgzi/SBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"964eef4d783d2739a99c9b357965467cc56789c6ac99c87f31ea4fb80152696e","last_reissued_at":"2026-08-13T01:30:13.564235Z","signature_status":"signed_v1","first_computed_at":"2026-08-13T01:30:13.564235Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RoutePack: Expert Placement and Attention-Aware Data Packing for MoE Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Gen Li, Xiaowei Zhu, Xudong Han, Yibo Shen, Zhenxuan Pan","submitted_at":"2026-08-12T15:07:15Z","abstract_excerpt":"Training Mixture-of-Experts (MoE) models for reinforcement learning (RL) couples two load-balancing problems: sequence composition determines dense attention work in each data-parallel microbatch, while token routing determines sparse expert work on expert-parallel ranks. Optimizing either alone can shift the bottleneck to the other. In MoE RL, rollout-time routing replay exposes every sample's sequence length and layer-wise expert demand before its training step. We present RoutePack, a hierarchical planner that coordinates state-consistent, layer-wise expert rerouting with joint attention- a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.12146","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/2608.12146/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":"2608.12146","created_at":"2026-08-13T01:30:13.565447+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.12146v1","created_at":"2026-08-13T01:30:13.565447+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.12146","created_at":"2026-08-13T01:30:13.565447+00:00"},{"alias_kind":"pith_short_12","alias_value":"SZHO6TLYHUTT","created_at":"2026-08-13T01:30:13.565447+00:00"},{"alias_kind":"pith_short_16","alias_value":"SZHO6TLYHUTTTKM4","created_at":"2026-08-13T01:30:13.565447+00:00"},{"alias_kind":"pith_short_8","alias_value":"SZHO6TLY","created_at":"2026-08-13T01:30:13.565447+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/SZHO6TLYHUTTTKM4TM2XSZKGPT","json":"https://pith.science/pith/SZHO6TLYHUTTTKM4TM2XSZKGPT.json","graph_json":"https://pith.science/api/pith-number/SZHO6TLYHUTTTKM4TM2XSZKGPT/graph.json","events_json":"https://pith.science/api/pith-number/SZHO6TLYHUTTTKM4TM2XSZKGPT/events.json","paper":"https://pith.science/paper/SZHO6TLY"},"agent_actions":{"view_html":"https://pith.science/pith/SZHO6TLYHUTTTKM4TM2XSZKGPT","download_json":"https://pith.science/pith/SZHO6TLYHUTTTKM4TM2XSZKGPT.json","view_paper":"https://pith.science/paper/SZHO6TLY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.12146&json=true","fetch_graph":"https://pith.science/api/pith-number/SZHO6TLYHUTTTKM4TM2XSZKGPT/graph.json","fetch_events":"https://pith.science/api/pith-number/SZHO6TLYHUTTTKM4TM2XSZKGPT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SZHO6TLYHUTTTKM4TM2XSZKGPT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SZHO6TLYHUTTTKM4TM2XSZKGPT/action/storage_attestation","attest_author":"https://pith.science/pith/SZHO6TLYHUTTTKM4TM2XSZKGPT/action/author_attestation","sign_citation":"https://pith.science/pith/SZHO6TLYHUTTTKM4TM2XSZKGPT/action/citation_signature","submit_replication":"https://pith.science/pith/SZHO6TLYHUTTTKM4TM2XSZKGPT/action/replication_record"}},"created_at":"2026-08-13T01:30:13.565447+00:00","updated_at":"2026-08-13T01:30:13.565447+00:00"}