ForeMoE uses routing foresight from the rollout stage to enable micro-step load balancing in MoE RL post-training via a hierarchical planner and transfer engine, claiming up to 1.45x speedup on 64 GPUs.
hub
Asyncflow: An asynchronous streaming rl framework for efficient llm post-training
18 Pith papers cite this work. Polarity classification is still indexing.
18
Pith papers citing it
hub tools
citation-role summary
background 2
citation-polarity summary
roles
background 2polarities
background 2representative citing papers
AstraFlow decouples RL components into autonomous dataflow services to natively support multi-policy agentic LLM training, elastic scaling, and cross-region execution with 2.7x speedup on math, code, search, and AgentBench workloads.
ReCrit frames critic interaction as a correctness-transition problem and uses quadrant-based RL rewards to improve LLM performance on scientific reasoning benchmarks by rewarding corrections and robustness while penalizing sycophancy.
HetRL delivers up to 9.17x higher throughput for LLM RL training on heterogeneous GPUs by using hybrid and ILP-based schedulers to solve a joint optimization problem over computation and data dependencies.
Complete-group pipelining plus frontier-group admission overlaps fixed-weight GRPO rollout and training, reducing wall-clock time 30.7–42.3% and trainer idle 37–76% versus Slime on Qwen3-1.7B.
Rollout-level advantage-prioritized experience replay for GRPO recycles high-advantage individual rollouts with age eviction and fresh-anchored batches to outperform standard GRPO on math benchmarks, with gains increasing with model size.
Mu-GRPO enables substantially more off-policy GRPO training for LLMs via relaxed clipping and negative-advantage veto in large staged batches, matching standard GRPO performance at ~2x training speed.
ROSE is a system for cooperative elasticity that co-locates serving and rollout models on shared GPUs, delivering 1.3-3.3x higher end-to-end throughput than fixed-resource baselines while preserving serving SLOs.
DORA's multi-version streaming rollout enables 2-3x higher throughput in asynchronous RL for LLMs while preserving convergence by maintaining policy consistency, data integrity, and bounded staleness.
Relax is a new RL training engine with omni-native design and async execution that delivers up to 2x speedups over baselines like veRL while converging to equivalent reward levels on Qwen3 models.
Seer improves synchronous LLM RL rollout throughput by up to 2.04x and reduces long-tail latency by 72-94% via divided rollout, context-aware scheduling, and adaptive grouped speculative decoding based on prompt similarity observations.
RLBoost harvests preemptible GPUs for RL rollout via a hybrid architecture with adaptive offload, pull-based transfer, and token-level migration, delivering 1.51x-1.97x throughput and 28-49% better cost efficiency than on-demand-only setups.
DARTS accelerates LLM RL training up to 1.77x by distribution-aware trajectory sampling and adaptive redundancy allocation that shapes rollouts toward conciseness without performance loss.
MinT is a system for managing million-scale LoRA adapter catalogs on shared 1T-parameter base models, with reported efficiency gains in adapter movement, multi-policy training, and catalog addressability.
Current agentic RL systems lack three key components needed for self-evolving agents at scale, requiring new co-designed architectures such as AReaL2.0 to enable policy updates from deployed workloads.
Xiaomi-GUI-0 reports 72.0% success on RealMobile and 78.9% on AndroidWorld via real-device closed-loop training with multi-source data and three-stage RL pipeline.
Libra optimizes GPU allocation across rollout and training in agentic RL via an elastic hybrid pool and C-MLFQ scheduler based on tool-return causal signals, claiming up to 3.0x throughput and 2.5x faster reward convergence on 48 A800 GPUs.
citing papers explorer
-
Harnessing Routing Foresight for Micro-step-level MoE load balancing in RL Post-training
ForeMoE uses routing foresight from the rollout stage to enable micro-step load balancing in MoE RL post-training via a hierarchical planner and transfer engine, claiming up to 1.45x speedup on 64 GPUs.
-
AstraFlow: Dataflow-Oriented Reinforcement Learning for Agentic LLMs
AstraFlow decouples RL components into autonomous dataflow services to natively support multi-policy agentic LLM training, elastic scaling, and cross-region execution with 2.7x speedup on math, code, search, and AgentBench workloads.
-
ReCrit: Transition-Aware Reinforcement Learning for Scientific Critic Reasoning
ReCrit frames critic interaction as a correctness-transition problem and uses quadrant-based RL rewards to improve LLM performance on scientific reasoning benchmarks by rewarding corrections and robustness while penalizing sycophancy.
-
HetRL: Efficient Reinforcement Learning for LLMs in Heterogeneous Environments
HetRL delivers up to 9.17x higher throughput for LLM RL training on heterogeneous GPUs by using hybrid and ILP-based schedulers to solve a joint optimization problem over computation and data dependencies.
-
RolloutPipe: Overlapping Pipelined Rollout and Training in Disaggregated On-Policy LLM Reinforcement Learning
Complete-group pipelining plus frontier-group admission overlaps fixed-weight GRPO rollout and training, reducing wall-clock time 30.7–42.3% and trainer idle 37–76% versus Slime on Qwen3-1.7B.
-
Rollout-Level Advantage-Prioritized Experience Replay for GRPO
Rollout-level advantage-prioritized experience replay for GRPO recycles high-advantage individual rollouts with age eviction and fresh-anchored batches to outperform standard GRPO on math benchmarks, with gains increasing with model size.
-
How Off-Policy Can GRPO Be? Mu-GRPO for Efficient LLM Reinforcement Learning
Mu-GRPO enables substantially more off-policy GRPO training for LLMs via relaxed clipping and negative-advantage veto in large staged batches, matching standard GRPO performance at ~2x training speed.
-
ROSE: Rollout On Serving GPUs via Cooperative Elasticity for Agentic RL
ROSE is a system for cooperative elasticity that co-locates serving and rollout models on shared GPUs, delivering 1.3-3.3x higher end-to-end throughput than fixed-resource baselines while preserving serving SLOs.
-
DORA: A Scalable Asynchronous Reinforcement Learning System for Language Model Training
DORA's multi-version streaming rollout enables 2-3x higher throughput in asynchronous RL for LLMs while preserving convergence by maintaining policy consistency, data integrity, and bounded staleness.
-
Relax: An Asynchronous Reinforcement Learning Engine for Omni-Modal Post-Training at Scale
Relax is a new RL training engine with omni-native design and async execution that delivers up to 2x speedups over baselines like veRL while converging to equivalent reward levels on Qwen3 models.
-
Seer: Online Context Learning for Fast Synchronous LLM Reinforcement Learning
Seer improves synchronous LLM RL rollout throughput by up to 2.04x and reduces long-tail latency by 72-94% via divided rollout, context-aware scheduling, and adaptive grouped speculative decoding based on prompt similarity observations.
-
RLBoost: Harvesting Preemptible Resources for Cost-Efficient Reinforcement Learning on LLMs
RLBoost harvests preemptible GPUs for RL rollout via a hybrid architecture with adaptive offload, pull-based transfer, and token-level migration, delivering 1.51x-1.97x throughput and 28-49% better cost efficiency than on-demand-only setups.
-
DARTS: Distribution-Aware Active Rollout Trajectory Shaping for Accelerating LLM Reinforcement Learning
DARTS accelerates LLM RL training up to 1.77x by distribution-aware trajectory sampling and adaptive redundancy allocation that shapes rollouts toward conciseness without performance loss.
-
MinT: Managed Infrastructure for Training and Serving Millions of LLMs
MinT is a system for managing million-scale LoRA adapter catalogs on shared 1T-parameter base models, with reported efficiency gains in adapter movement, multi-policy training, and catalog addressability.
-
Next-Generation Agentic Reinforcement Learning Systems Enable Self-Evolving Agents
Current agentic RL systems lack three key components needed for self-evolving agents at scale, requiring new co-designed architectures such as AReaL2.0 to enable policy updates from deployed workloads.
-
Xiaomi-GUI-0 Technical Report
Xiaomi-GUI-0 reports 72.0% success on RealMobile and 78.9% on AndroidWorld via real-device closed-loop training with multi-source data and three-stage RL pipeline.
-
Libra: Efficient Resource Management for Agentic RL Post-Training
Libra optimizes GPU allocation across rollout and training in agentic RL via an elastic hybrid pool and C-MLFQ scheduler based on tool-return causal signals, claiming up to 3.0x throughput and 2.5x faster reward convergence on 48 A800 GPUs.
- AsyncWebRL: Efficient Asynchronous Reinforcement Learning for Multi-Step Visual Web Agents