DORA: A Scalable Asynchronous Reinforcement Learning System for Language Model Training

• URL: http://arxiv.org/abs/2604.26256v1
• Date: Wed, 29 Apr 2026 03:25:36 GMT
• Title: DORA: A Scalable Asynchronous Reinforcement Learning System for Language Model Training
• Abstract summary: Asynchronous training offers a natural remedy by overlapping generation with training.<n>Existing approaches fail to intrinsically address the long-tailed trajectory problem.<n>We propose DORA, which addresses this challenge through algorithm-system co-design.
• Score: 30.35748518368946
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Reinforcement learning (RL) has become a critical paradigm for LLM post-training, yet the rollout phase -- accounting for 50--80% of total step time -- is bottlenecked by skewed generation: long-tailed trajectories indispensable for model performance block the entire training pipeline. Asynchronous training offers a natural remedy by overlapping generation with training, but introduces a fundamental tension between efficiency and algorithmic correctness. We identify three constraints in asynchronous training to preserve convergence: intra-trajectory policy consistency, data integrity, and bounded staleness. Existing approaches fail to intrinsically address the long-tailed trajectory problem, which is further exacerbated by the imbalance characteristic of Mix-of-Experts models, or deviate from the standard RL training formulation, thereby hindering model convergence. Therefore, we propose DORA (Dynamic ORchestration for Asynchronous Rollout), which addresses this challenge through algorithm-system co-design. DORA introduces multi-version streaming rollout, a novel asynchronous paradigm that maintains multiple policy versions concurrently -- simultaneously achieving full bubble elimination without compromising algorithmic constraints. Experimental results demonstrate that our DORA system achieves substantial improvements in throughput -- up to 2--3 times higher than state-of-the-art systems on open-source benchmarks -- without compromising convergence. Furthermore, in large-scale industrial applications with tens of thousands of accelerators, DORA accelerates RL training by 2--4 times compared to synchronous training across various scenarios. The resultant open-source models, LongCat-Flash-Thinking, exhibit competitive performance on complex reasoning benchmarks, matching the capability of most advanced LLMs.

Related papers

• Single-Rollout Asynchronous Optimization for Agentic Reinforcement Learning [47.00572734047258]
  We present Single-rollout Asynchronous Optimization (SAO) to address the stability and off-policy challenges in asynchronous RL.<n>SAO replaces group-wise sampling with single-rollout sampling, that is, using one rollout per prompt.<n>SAO is successfully deployed in the agentic RL pipeline for training the open GLM-5.2 model.
  arXiv  Detail & Related papers  (2026-07-08T15:02:19Z)
• SortedRL: Accelerating RL Training for LLMs through Online Length-Aware Scheduling [54.276306194000405]
  We propose SortedRL, an online length-aware scheduling strategy for scaling reinforcement learning.<n>SortedRL reorders rollout samples based on output lengths, prioritizing short samples forming groups for early updates.<n>Experiments using LLaMA-3.1-8B and Qwen-2.5-32B on diverse tasks, including logical puzzles, show that SortedRL reduces RL training bubble ratios by over 50%.
  arXiv  Detail & Related papers  (2026-03-24T16:48:31Z)
• AcceRL: A Distributed Asynchronous Reinforcement Learning and World Model Framework for Vision-Language-Action Models [21.204182133307047]
  We propose AcceRL, a fully asynchronous and decoupled RL framework.<n>AcceRL is the first to integrate a plug-and-play, trainable world model into a distributed asynchronous RL pipeline.<n> Experiments on the LIBEROciteliu2023libero benchmark demonstrate that AcceRL achieves state-of-the-art (SOTA) performance.
  arXiv  Detail & Related papers  (2026-03-19T03:50:45Z)
• RL-VLA$^3$: Reinforcement Learning VLA Accelerating via Full Asynchronism [42.27384804295299]
  Vision-Language-Action (VLA) models have emerged as a crucial pathway towards general embodied intelligence.<n>This paper proposes and implements a fully-asynchronous policy training framework encompassing the entire pipeline from environment interaction to actor policy updates.<n>On the LIBERO benchmark, the framework achieves throughput improvements of up to 59.25% compared to existing synchronous strategies.
  arXiv  Detail & Related papers  (2026-02-05T15:30:23Z)
• DiRL: An Efficient Post-Training Framework for Diffusion Language Models [54.405206032785706]
  Diffusion Language Models (dLLMs) have emerged as promising alternatives to Auto-Regressive (AR) models.<n>Existing methods suffer from computational inefficiency and objective mismatches between training and inference.<n>We introduce DiRL, an efficient post-training framework that tightly integrates FlexAttention-accelerated blockwise training with LMDeploy-optimized inference.
  arXiv  Detail & Related papers  (2025-12-23T08:33:19Z)
• Periodic Asynchrony: An Effective Method for Accelerating Reinforcement Learning [8.395046547177806]
  reinforcement learning (RL) has attracted increasing attention, with growing efforts to reproduce and apply it.<n>In mainstream RL frameworks, inference and training are typically deployed on the same devices.<n>In this study, we are returning to the strategy of separating inference and training deployment.<n>We transform the conventional synchronous architecture into a periodically asynchronous framework, which allows for demand-driven, independent, and elastic scaling of each component.
  arXiv  Detail & Related papers  (2025-11-24T08:22:50Z)
• Taming the Long-Tail: Efficient Reasoning RL Training with Adaptive Drafter [52.111923076688505]
  Training Large Language Models (LLMs) with strong reasoning capabilities marks a significant milestone, unlocking new frontiers in complex problem-solving.<n>We propose TLT, a system that accelerates reasoning RL training losslessly by integrating adaptive speculative decoding.
  arXiv  Detail & Related papers  (2025-11-20T18:59:25Z)
• AReaL-Hex: Accommodating Asynchronous RL Training over Heterogeneous GPUs [24.96730768606278]
  We present AReaL-Hex, a heterogeneous-aware asynchronous RL training system.<n>It effectively schedules how to execute rollout generation and policy model training over heterogeneous GPUs.<n>It delivers up to 1.50x higher training throughput and 1.46x reduction in training cost.
  arXiv  Detail & Related papers  (2025-11-02T04:17:30Z)
• Laminar: A Scalable Asynchronous RL Post-Training Framework [20.127034898123508]
  Long-tail skewness in RL trajectory generation causes severe GPU underutilization.<n>Current RL systems rely on global weight synchronization between the actor and all rollouts, which creates a rigid model update schedule.<n>We propose Laminar, a scalable and robust RL post-training system built on a fully decoupled architecture.
  arXiv  Detail & Related papers  (2025-10-14T15:29:14Z)
• Part II: ROLL Flash -- Accelerating RLVR and Agentic Training with Asynchrony [78.70328630805041]
  ROLL Flash is a system that extends ROLL with native support for asynchronous RL post-training.<n>We show that ROLL Flash significantly improves resource utilization and scalability over synchronous RL post-training.
  arXiv  Detail & Related papers  (2025-10-13T12:41:27Z)
• Ring-lite: Scalable Reasoning via C3PO-Stabilized Reinforcement Learning for LLMs [51.21041884010009]
  Ring-lite is a Mixture-of-Experts (MoE)-based large language model optimized via reinforcement learning (RL)<n>Our approach matches the performance of state-of-the-art (SOTA) small-scale reasoning models on challenging benchmarks.
  arXiv  Detail & Related papers  (2025-06-17T17:12:34Z)
• StreamRL: Scalable, Heterogeneous, and Elastic RL for LLMs with Disaggregated Stream Generation [55.75008325187133]
  Reinforcement learning (RL) has become the core post-training technique for large language models (LLMs)<n>StreamRL is designed with disaggregation from first principles to address two types of performance bottlenecks.<n> Experiments show that StreamRL improves throughput by up to 2.66x compared to existing state-of-the-art systems.
  arXiv  Detail & Related papers  (2025-04-22T14:19:06Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.