{"paper":{"title":"A Readiness-Driven Runtime for Pipeline-Parallel Training under Runtime Variability","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.DC","authors_text":"(2) ScitiX AI), Alan Zhao (2), Guang Yang (1), Ruitao Liu (1), Shuo Chen (1), Tingrui Zhang (1), Wei Xu (1) ((1) Tsinghua University, Xinyang Tian (1)","submitted_at":"2026-05-18T17:59:18Z","abstract_excerpt":"Pipeline parallelism is a key technique for scaling large-model training, but modern workloads exhibit runtime variability in computation and communication. Existing pipeline systems typically consume static, profiled, or adaptively generated schedules as pre-committed execution orders. When realized task readiness diverges from the pre-committed order, stages may wait for not-yet-ready work even though other executable work is available, creating stage misalignment, idle bubbles, and reduced utilization.\n  We present Runtime-Readiness-First Pipeline (RRFP), a readiness-driven runtime for pipe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.18750","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/2605.18750/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T00:01:58.976906Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"62b623c2768752ac9d6310457c913a52247ec08e46d88c601306c43c5840dbd1"},"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"}