cs.AINov 4, 2026

AdaFloor: Adaptive Floor Planning and Rank-Time Reuse for Long-Tail MoE RL Rollouts

Authors: Jia Wei, Ziyi Qiu, Yi Xu, Xuehai Qian

Organizations: Department of Computer Science and Technology, Tsinghua University

Abstract

Synchronous MoE reinforcement-learning post-training suffers from response-length imbalance. Existing methods reduce request tails, yet completed ranks remain bound to Expert Parallelism (EP) collectives and expert ownership. In a representative 16-accelerator Qwen3-30B-A3B rollout, this dummy run traps 32.83% of rank-time. AdaFloor converts this EP-coupled capacity into bounded leases for auxiliary work to improve system resource efficiency through rank-time reuse. Historical prompt-tail prediction and a calibrated guard construct the KV-safe layout. AdaFloor then selects the floor-specific safe quorum schedule with the largest predicted release area. Its runtime shrinks the EP group and leases detached ranks until full-world restoration. Across three updated-policy seeds on DeepScaleR mathematical reasoning rollouts, its capacity-oriented Natural floor2 configuration with TailGuard reduces Vanilla's 29393.9 rank-seconds of dummy run by 96.15%. Rollout shortening eliminates 18693.6 rank-seconds, while coordinated exits convert another 9568.6 rank-seconds into reusable capacity, leaving 1131.7 rank-seconds. For sidecar reuse, its Planned residency floor4 configuration without TailGuard serves 3134 Qwen3-8B GSM8K queries per epoch at 99.46% accuracy and incurs only a 2 to 3% primary throughput reduction against the same configuration with sidecar execution disabled. Its latency-oriented Planned residency floor4 configuration with TailGuard reduces rollout time by 24.09% relative to Vanilla and by 2.31% relative to LengthSort with TailGuard. On public Eurus-2-RL code rollouts with the same model and 16-rank topology, AdaFloor reduces rollout time by 20.56% and dummy run by 89.13% while generated work and mean reward remain stable.

Figures & tables

Explore similar work

Oct 15, 2026

ReFlow: Replay-Driven Decoupling for Efficient RL Post-Training

Reinforcement learning (RL) has been a critical paradigm for improving large language model (LLM) reasoning and agentic capabilities, but its efficiency remains a major bottleneck. Existing RL systems commonly rely on asynchronous execution to overlap rollout generation with policy optimization for enhanced efficiency. However, this approach faces inherent limitations: mild policy staleness offers only limited training efficiency gains, whereas aggressive asynchrony can degrade model quality and yields diminishing efficiency gains. The dynamic nature of LLM RL and multi-turn agentic workloads further challenges RL system design, as resource demands shift over time, making static resource allocation and generation strategy inefficient. This paper presents ReFlow, a system-algorithm co-designed RL framework for efficient LLM post-training. Its key insight is that, unlike existing RL frameworks which consume each trajectory only once, experience replay of recent trajectories enables effective sample reuse, thereby amortizing the high cost of rollout generation without incurring excessive policy staleness. To fully exploit this opportunity, ReFlow adopts a decoupled and adaptive architecture with three key designs: a dual replay data cycle supports efficient experience reuse while balancing training throughput and optimization stability; an elastic resource pool enables dynamic resource placement to accommodate evolving RL workloads; an adaptive scheduler coordinates generation strategies and resource allocation to improve resource utilization and boost generation throughput. Experiments across representative single-turn reasoning and multi-turn agentic RL tasks show that ReFlow improves training throughput by up to 2.06×2.06\times over state-of-the-art synchronous and asynchronous RL systems, while maintaining comparable or better model performance.
Nov 6, 2026

Adaptive Scheduling Pipeline for Multi-Instance Asynchronous Reinforcement Learning

We present the Adaptive Scheduling Pipeline (ASP), an offline rollout scheduler for multi-instance asynchronous reinforcement learning (RL) of large language models (LLMs). ASP jointly reduces three sources of rollout underutilization - padding bubbles from intra-minibatch length heterogeneity, skewness bubbles from long-tail outputs, and sorting bubbles from fluctuations in training-queue occupancy induced by output-length-sorted dispatch - through a single offline preprocessing pass. This enables efficient multi-instance rollout that overcomes the early scaling plateau of single-instance inference while reducing fluctuations in the supply of generated samples to the trainer. We integrate the scheduler into an asynchronous architecture that decouples rollout from training while keeping policy lag observable and bounded. Within our asynchronous architecture, ASP's scheduling reduces mean iteration time by 10.9%. The complete system improves end-to-end throughput by up to 3.22 times and reduces iteration time by up to 76.8% relative to MindSpeed-RL, while reducing iteration time by up to 62.5% relative to verl. It also scales more robustly with additional generation resources than these baselines and shows no visible convergence degradation in the measured bounded-lag regime. ASP has since been used in an internal telecom-domain RL post-training workflow, reducing the reported training period from more than three months to two months and producing a domain-enhanced model for knowledge answering and tool calling.
Aug 11, 2026

TideRL: Boosting Agentic RL Goodput with Readiness-Aware Scheduling

Reinforcement learning (RL) for large language models is moving toward multi-turn agentic workloads, where rollout tasks repeatedly pause for external environments, resume with growing contexts, and finish at highly variable times. In this setting, RL training goodput, measured by training throughput, matters more than raw GPU occupancy: GPU waiting and repeated prefill recomputation are pure overhead. We present TideRL, a readiness-aware elastic RL system with Continuous Task Batching, Resource-Aware Ref-Actor Pipelining, and Elastic Resource Scaling. CTB preserves useful rollout state, RA2P\textrm{RA}^2\textrm{P} selects between decoupled streaming and colocated aggregation from the ready backlog and arrival interval, and ERS moves ranks between rollout and training using the same readiness signals. Across text-only and multi-modal agentic workloads, TideRL improves RL training goodput by up to 5.6×\times over synchronous baselines and over 33% over asynchronous baselines, while reaching similar task performance. It also improves KV cache hit rate by 1.58×\times, reduces per-step training time by up to 44.3%, and cuts total waiting time by up to 77.6%.