# A-3PO: Accelerating Asynchronous LLM Training with Staleness-aware Proximal Policy Approximation

arXiv: https://arxiv.org/abs/2512.06547 · HTML: https://mlsanity.com/paper/2512.06547/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Builds on 29 of its 36 references that are in the map.
- First posted on arXiv 6 Dec 2025.

## Builds on (most cited first, 29 of 29)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [DAPO: An Open-Source LLM Reinforcement Learning System at Scale](https://mlsanity.com/paper/2503.14476/) (2025)
- [Trust Region Policy Optimization](https://mlsanity.com/paper/1502.05477/) (2015)
- [HybridFlow: A Flexible and Efficient RLHF Framework](https://mlsanity.com/paper/2409.19256/) (2024)
- [SGLang: Efficient Execution of Structured Language Model Programs](https://mlsanity.com/paper/2312.07104/) (2023)
- [IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures](https://mlsanity.com/paper/1802.01561/) (2018)
- [Group Sequence Policy Optimization](https://mlsanity.com/paper/2507.18071/) (2025)
- [Safe RLHF: Safe Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2310.12773/) (2023)
- [RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback](https://mlsanity.com/paper/2309.00267/) (2023)
- [RLHF Workflow: From Reward Modeling to Online RLHF](https://mlsanity.com/paper/2405.07863/) (2024)
- [A Survey of Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2312.14925/) (2023)
- [AReaL: A Large-Scale Asynchronous Reinforcement Learning System for Language Reasoning](https://mlsanity.com/paper/2505.24298/) (2025)
- [Secrets of RLHF in Large Language Models Part II: Reward Modeling](https://mlsanity.com/paper/2401.06080/) (2024)
- [Reinforcement Learning Optimization for Large-Scale Learning: An Efficient and User-Friendly Scaling Library](https://mlsanity.com/paper/2506.06122/) (2025)
- [Asynchronous RLHF: Faster and More Efficient Off-Policy RL for Language Models](https://mlsanity.com/paper/2410.18252/) (2024)
- [LlamaRL: A Distributed Asynchronous Reinforcement Learning Framework for Efficient Large-scale LLM Training](https://mlsanity.com/paper/2505.24034/) (2025)
- [Batch size-invariance for policy optimization](https://mlsanity.com/paper/2110.00641/) (2021)
- [Efficient Training of Large Language Models on Distributed Infrastructures: A Survey](https://mlsanity.com/paper/2407.20018/) (2024)
- [Asynchronous Federated Reinforcement Learning with Policy Gradient Updates: Algorithm Design and Convergence Analysis](https://mlsanity.com/paper/2404.08003/) (2024)
- [Soft Policy Optimization: Online Off-Policy RL for Sequence Models](https://mlsanity.com/paper/2503.05453/) (2025)
- [Policy Gradient with Active Importance Sampling](https://mlsanity.com/paper/2405.05630/) (2024)
- [MindSpeed RL: Distributed Dataflow for Scalable and Efficient RL Training on Ascend NPU Cluster](https://mlsanity.com/paper/2507.19017/) (2025)
- [Value-aware Importance Weighting for Off-policy Reinforcement Learning](https://mlsanity.com/paper/2306.15625/) (2023)

## Built on by (most cited first, 12 of 12)

- [Open-MOPD: Diagnosing and Fixing Capability Imbalance in Multi-Teacher On-Policy Distillation](https://mlsanity.com/paper/2608.19098/) (2026)
- [GAC: Stabilizing Asynchronous RL Training for LLMs via Gradient Alignment Control](https://mlsanity.com/paper/2603.01501/) (2026)
- [Missing Old Logits in Asynchronous Agentic RL: Semantic Mismatch and Repair Methods for Off-Policy Correction](https://mlsanity.com/paper/2605.12070/) (2026)
- [AsyncOPD: How Stale Can On-Policy Distillation Be?](https://mlsanity.com/paper/2606.24143/) (2026)
- [RLHFless: Serverless Computing for Efficient RLHF](https://mlsanity.com/paper/2602.22718/) (2026)
- [Staleness-Learning Rate Scaling Laws for Asynchronous RLHF](https://mlsanity.com/paper/2607.01083/) (2026)
- [SPO++: Stream-Aligned Policy Optimization for Asynchronous Agentic RL](https://mlsanity.com/paper/2608.24870/) (2026)
- [Local Guidance, Global Impact: Gaussian-Reshaped Trust Region Unlocks Behavior Transitions](https://mlsanity.com/paper/2606.03382/) (2026)
- [Retroactive Advantage Correction: Closed-Form V-Trace Bias Correction for Delay-Aware RLHF](https://mlsanity.com/paper/2606.27580/) (2026)
- [Cool the Sampler, Not the Learner: Sampling Temperature Moves the Staleness Cliff of Importance-Corrected GRPO](https://mlsanity.com/paper/2609.36953/) (2026)
- [Asynchronous LLM Post-Training: Group-Mass Capping and Convergence Analysis](https://mlsanity.com/paper/2610.01896/) (2026)
- [Probe the Harness: Setup Checks for Stale-Data RL Comparisons in Language Models](https://mlsanity.com/paper/2610.02911/) (2026)

## Abstract (arXiv)

Decoupled PPO has been a successful reinforcement learning (RL) algorithm to deal with the high data staleness under the asynchronous RL setting. Decoupled loss used in decoupled PPO improves coupled-loss style of algorithms' (e.g., standard PPO, GRPO) learning stability by introducing a proximal policy to decouple the off-policy correction (importance weight) from the policy update constraint (trust region). However, the proximal policy requires an extra forward pass through the model at each training step, creating a computational overhead for large language models training. We observe that since the proximal policy only serves as a trust region anchor between the behavior and target policies, we can approximate it through simple interpolation without explicit computation. We call this approach A-3PO (APproximated Proximal Policy Optimization). A-3PO eliminates this overhead, accelerating training by 1.8x speedup while maintaining comparable performance. Code & off-the-shelf example are contributed to the open-source RL training system AReaL at: https://github.com/areal-project/AReaL/blob/v1.0.0.rc1/docs/algorithms/proxapprox.md