Learning to Act While Waiting: RL Finetuning of Generalist Robot Policies Under Inference Latency

• URL: http://arxiv.org/abs/2608.23831v2
• Date: Wed, 26 Aug 2026 14:32:04 GMT
• Title: Learning to Act While Waiting: RL Finetuning of Generalist Robot Policies Under Inference Latency
• Abstract summary: We introduce a latency-aware framework, Asynchronous RL with Intermediate Information (ARLI), that enables RL-based improvement of generalist policies under inference delays.<n>Our framework builds on asynchronous inference approaches, which interleave action generation with execution to hide latency, and addresses its incompatibility with RL.<n>We evaluate our approach across simulated and real-world manipulation tasks, and find that it enables effective finetuning under inference delays where standard RL fails entirely.
• Score: 38.99457938189648
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: While reinforcement learning (RL) allows generalist robot policies to continually improve during deployment, the large model size of modern generalist policies, such as VLAs, poses a fundamental obstacle to effective RL improvement. In particular, their severe inference latency---which can lead to pauses or jerky movements---can alter the effective environment dynamics and, if not correctly accounted for, break the Markov assumption that RL relies on, causing standard RL algorithms to fail completely. In this work, we introduce a latency-aware framework, Asynchronous RL with Intermediate Information (ARLI), that enables RL-based improvement of generalist policies under inference delays. Our framework builds on asynchronous inference approaches, which interleave action generation with execution to hide latency, and addresses its incompatibility with RL by providing a low-latency RL policy design that maximizes reactivity within the inference window through two contributions: state augmentations that restore near-Markovian structure by incorporating committed actions and a mid-inference observation. We evaluate our approach across simulated and real-world manipulation tasks, and find that it enables effective finetuning under inference delays where standard RL fails entirely, even matching or exceeding the performance of standard RL in idealized no-latency settings.

Related papers

• SmoothRL: Online Reinforcement Learning During Asynchronous Execution [12.24698147742807]
  Deploying robot policies in the physical world requires satisfying two fundamental desiderata: reliability and smooth real-time execution.<n>We present SmoothRL, an online reinforcement learning framework that fine-tunes a pretrained policy within an asynchronous inference loop.<n>We evaluate SmoothRL on real-world robotic tasks requiring high precision, as well as highly dynamic tasks that necessitate asynchronous execution.
  arXiv  Detail & Related papers  (2026-08-30T12:56:09Z)
• Scaling Automatic Research Agents via World Models [42.9808933592206]
  Recent research (AutoResearch) agents show the capability to independently implement solutions and learn from the execution.<n>Behind these gains, post-training (especially RL) plays a central role.<n>World Model RL (WMRL) replaces environment execution with a world model to remove this bottleneck.
  arXiv  Detail & Related papers  (2026-08-12T20:11:25Z)
• Found-RL: foundation model-enhanced reinforcement learning for autonomous driving [15.275134927543611]
  Reinforcement Learning (RL) has emerged as a dominant paradigm for end-to-end autonomous driving (AD)<n>Found-RL is a platform tailored to efficiently enhance RL for AD using foundation models.<n>A core innovation is the asynchronous batch inference framework, which decouples heavy VLM reasoning from the simulation loop.
  arXiv  Detail & Related papers  (2026-02-11T02:56:04Z)
• REBEL: Reinforcement Learning via Regressing Relative Rewards [59.68420022466047]
  We propose REBEL, a minimalist RL algorithm for the era of generative models.<n>In theory, we prove that fundamental RL algorithms like Natural Policy Gradient can be seen as variants of REBEL.<n>We find that REBEL provides a unified approach to language modeling and image generation with stronger or similar performance as PPO and DPO.
  arXiv  Detail & Related papers  (2024-04-25T17:20:45Z)
• Boosting Reinforcement Learning with Strongly Delayed Feedback Through Auxiliary Short Delays [41.52768902667611]
  Reinforcement learning (RL) is challenging in the common case of delays between events and their sensory perceptions. We present a novel Auxiliary-Delayed Reinforcement Learning (AD-RL) method that leverages auxiliary tasks involving short delays to accelerate RL with long delays. Specifically, AD-RL learns a value function for short delays and uses bootstrapping and policy improvement techniques to adjust it for long delays.
  arXiv  Detail & Related papers  (2024-02-05T16:11:03Z)
• Hybrid Reinforcement Learning for Optimizing Pump Sustainability in Real-World Water Distribution Networks [55.591662978280894]
  This article addresses the pump-scheduling optimization problem to enhance real-time control of real-world water distribution networks (WDNs) Our primary objectives are to adhere to physical operational constraints while reducing energy consumption and operational costs. Traditional optimization techniques, such as evolution-based and genetic algorithms, often fall short due to their lack of convergence guarantees.
  arXiv  Detail & Related papers  (2023-10-13T21:26:16Z)
• Mastering the Unsupervised Reinforcement Learning Benchmark from Pixels [112.63440666617494]
  Reinforcement learning algorithms can succeed but require large amounts of interactions between the agent and the environment. We propose a new method to solve it, using unsupervised model-based RL, for pre-training the agent. We show robust performance on the Real-Word RL benchmark, hinting at resiliency to environment perturbations during adaptation.
  arXiv  Detail & Related papers  (2022-09-24T14:22:29Z)
• Improving Zero-shot Generalization in Offline Reinforcement Learning using Generalized Similarity Functions [34.843526573355746]
  Reinforcement learning (RL) agents are widely used for solving complex sequential decision making tasks, but exhibit difficulty in generalizing to scenarios not seen during training. We show that performance of online algorithms for generalization in RL can be hindered in the offline setting due to poor estimation of similarity between observations. We propose a new theoretically-motivated framework called Generalized Similarity Functions (GSF), which uses contrastive learning to train an offline RL agent to aggregate observations based on the similarity of their expected future behavior.
  arXiv  Detail & Related papers  (2021-11-29T15:42:54Z)
• Behavioral Priors and Dynamics Models: Improving Performance and Domain Transfer in Offline RL [82.93243616342275]
  We introduce Offline Model-based RL with Adaptive Behavioral Priors (MABE) MABE is based on the finding that dynamics models, which support within-domain generalization, and behavioral priors, which support cross-domain generalization, are complementary. In experiments that require cross-domain generalization, we find that MABE outperforms prior methods.
  arXiv  Detail & Related papers  (2021-06-16T20:48:49Z)
• Progressive extension of reinforcement learning action dimension for asymmetric assembly tasks [7.4642148614421995]
  In this paper, a progressive extension of action dimension (PEAD) mechanism is proposed to optimize the convergence of RL algorithms. The results demonstrate the PEAD method will enhance the data-efficiency and time-efficiency of RL algorithms as well as increase the stable reward.
  arXiv  Detail & Related papers  (2021-04-06T11:48:54Z)
• Robust Deep Reinforcement Learning against Adversarial Perturbations on State Observations [88.94162416324505]
  A deep reinforcement learning (DRL) agent observes its states through observations, which may contain natural measurement errors or adversarial noises. Since the observations deviate from the true states, they can mislead the agent into making suboptimal actions. We show that naively applying existing techniques on improving robustness for classification tasks, like adversarial training, is ineffective for many RL tasks.
  arXiv  Detail & Related papers  (2020-03-19T17:59:59Z)

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.