Reinforcement Learning in Robotics

Últimos artículos

• Beyond Mode Elicitation: Diversity-Preserving Reinforcement Learning via Latent Diffusion Reasoner

  Haoqiang Kang, Yizhe Zhang, Nikki Lijing Kuang, Yi-An Ma, Lianhui Qin · 3 de febrero de 2026

  Recent reinforcement learning (RL) methods improve LLM reasoning by optimizing discrete Chain-of-Thought (CoT) generation; however, exploration in token space often suffers from diversity collapse as policy entropy decreases due to mode elicitation behavior in discrete RL. To mitigate this issue, we…

• World-Gymnast: Training Robots with Reinforcement Learning in a World Model

  Ansh Kumar Sharma, Yixiang Sun, Ninghao Lu, Yunzhe Zhang, Jiarao Liu, Sherry Yang · 3 de febrero de 2026

  Robot learning from interacting with the physical world is fundamentally bottlenecked by the cost of physical interaction. The two alternatives, supervised finetuning (SFT) from expert demonstrations and reinforcement learning (RL) in a software-based simulator, are limited by the amount of expert d…

• Good SFT Optimizes for SFT, Better SFT Prepares for Reinforcement Learning

  Dylan Zhang, Yufeng Xu, Haojin Wang, Qingzhi Chen, Hao Peng · 3 de febrero de 2026

  Post-training of reasoning LLMs is a holistic process that typically consists of an offline SFT stage followed by an online reinforcement learning (RL) stage. However, SFT is often optimized in isolation to maximize SFT performance alone. We show that, after identical RL training, models initializ…

• TABX: A High-Throughput Sandbox Battle Simulator for Multi-Agent Reinforcement Learning

  Hayeong Lee, JunHyeok Oh, Byung-Jun Lee · 3 de febrero de 2026

  The design of environments plays a critical role in shaping the development and evaluation of cooperative multi-agent reinforcement learning (MARL) algorithms. While existing benchmarks highlight critical challenges, they often lack the modularity required to design custom evaluation scenarios. We i…

• Online Fine-Tuning of Pretrained Controllers for Autonomous Driving via Real-Time Recurrent RL

  Julian Lemmel, Felix Resch, M\'onika Farsang, Ramin Hasani, Daniela Rus, Radu Grosu · 3 de febrero de 2026

  Deploying pretrained policies in real-world applications presents substantial challenges that fundamentally limit the practical applicability of learning-based control systems. When autonomous systems encounter environmental changes in system dynamics, sensor drift, or task objectives, fixed policie…

• Reasoning and Tool-use Compete in Agentic RL:From Quantifying Interference to Disentangled Tuning

  Yu Li, Mingyang Yi, Xiuyu Li, Ju Fan, Fuxin Jiang, Binbin Chen, Peng Li, Jie Song, Tieying Zhang · 3 de febrero de 2026

  Agentic Reinforcement Learning (ARL) focuses on training large language models (LLMs) to interleave reasoning with external tool execution to solve complex tasks. Most existing ARL methods train a single shared model parameters to support both reasoning and tool use behaviors, implicitly assuming th…

• Action-Free Offline-to-Online RL via Discretised State Policies

  Natinael Solomon Neggatu, Jeremie Houssineau, Giovanni Montana · 3 de febrero de 2026

  Most existing offline RL methods presume the availability of action labels within the dataset, but in many practical scenarios, actions may be missing due to privacy, storage, or sensor limitations. We formalise the setting of action-free offline-to-online RL, where agents must learn from datasets c…

• ESSAM: A Novel Competitive Evolution Strategies Approach to Reinforcement Learning for Memory Efficient LLMs Fine-Tuning

  Zhishen Sun, Sizhe Dang, Guang Dai, Haishan Ye · 3 de febrero de 2026

  Reinforcement learning (RL) has become a key training step for improving mathematical reasoning in large language models (LLMs), but it often has high GPU memory usage, which makes it hard to use in settings with limited resources. To reduce these issues, we propose Evolution Strategies with Sharpne…

• TQL: Scaling Q-Functions with Transformers by Preventing Attention Collapse

  Perry Dong, Kuo-Han Hung, Alexander Swerdlow, Dorsa Sadigh, Chelsea Finn · 3 de febrero de 2026

  Despite scale driving substantial recent advancements in machine learning, reinforcement learning (RL) methods still primarily use small value functions. Naively scaling value functions -- including with a transformer architecture, which is known to be highly scalable -- often results in learning in…

• Boosting Maximum Entropy Reinforcement Learning via One-Step Flow Matching

  Zeqiao Li, Yijing Wang, Haoyu Wang, Zheng Li, Zhiqiang Zuo · 3 de febrero de 2026

  Diffusion policies are expressive yet incur high inference latency. Flow Matching (FM) enables one-step generation, but integrating it into Maximum Entropy Reinforcement Learning (MaxEnt RL) is challenging: the optimal policy is an intractable energy-based distribution, and the efficient log-likelih…

• Probabilistic Performance Guarantees for Multi-Task Reinforcement Learning

  Yannik Schnitzer, Mathias Jackermeier, Alessandro Abate, David Parker · 3 de febrero de 2026

  Multi-task reinforcement learning trains generalist policies that can execute multiple tasks. While recent years have seen significant progress, existing approaches rarely provide formal performance guarantees, which are indispensable when deploying policies in safety-critical settings. We present a…

• ECHO: Entropy-Confidence Hybrid Optimization for Test-Time Reinforcement Learning

  Chu Zhao, Enneng Yang, Yuting Liu, Jianzhe Zhao, Guibing Guo · 3 de febrero de 2026

  Test-time reinforcement learning generates multiple candidate answers via repeated rollouts and performs online updates using pseudo-labels constructed by majority voting. To reduce overhead and improve exploration, prior work introduces tree structured rollouts, which share reasoning prefixes and b…

• Reinforcement Learning for Control Systems with Time Delays: A Comprehensive Survey

  Armando Alves Neto · 3 de febrero de 2026

  In the last decade, Reinforcement Learning (RL) has achieved remarkable success in the control and decision-making of complex dynamical systems. However, most RL algorithms rely on the Markov Decision Process assumption, which is violated in practical cyber-physical systems affected by sensing delay…

• World Models as an Intermediary between Agents and the Real World

  Sherry Yang · 3 de febrero de 2026

  Large language model (LLM) agents trained using reinforcement learning has achieved superhuman performance in low-cost environments like games, mathematics, and coding. However, these successes have not translated to complex domains where the cost of interaction is high, such as the physical cost of…

• TIDE: Trajectory-based Diagnostic Evaluation of Test-Time Improvement in LLM Agents

  Hang Yan, Xinyu Che, Fangzhi Xu, Qiushi Sun, Zichen Ding, Kanzhi Cheng, Jian Zhang, Tao Qin, Jun Liu, Qika Lin · 3 de febrero de 2026

  Recent advances in autonomous LLM agents demonstrate their ability to improve performance through iterative interaction with the environment. We define this paradigm as Test-Time Improvement (TTI). However, the mechanisms under how and why TTI succeed or fail remain poorly understood, and existing e…

• Learning Robust Reasoning through Guided Adversarial Self-Play

  Shuozhe Li, Vaishnav Tadiparthi, Kwonjoon Lee, Nakul Agarwal, Hossein Nourkhiz Mahjoub, Ehsan Moradi Pari, Lizhang Chen, Amy Zhang, Liu Leqi · 3 de febrero de 2026

  Reinforcement learning from verifiable rewards (RLVR) produces strong reasoning models, yet they can fail catastrophically when the conditioning context is fallible (e.g., corrupted chain-of-thought, misleading partial solutions, or mild input perturbations), since standard RLVR optimizes final-answ…

• Past-Discounting is Key for Learning Markovian Fairness with Long Horizons

  Ashwin Kumar, William Yeoh · 3 de febrero de 2026

  Fairness is an important consideration for dynamic resource allocation in multi-agent systems. Many existing methods treat fairness as a one-shot problem without considering temporal dynamics, which misses the nuances of accumulating inequalities over time. Recent approaches overcome this limitation…

• Evolving Interpretable Constitutions for Multi-Agent Simulation

  Ujwal Kumar, Alice Saito, Hershraj Niranjani, Rayan Yessou, Phan Xuan Tan · 3 de febrero de 2026

  Constitutional AI has focused on single-model alignment using fixed principles. However, multi-agent systems create novel alignment challenges through emergent social dynamics. We present Constitutional Evolution, a framework for automatically discovering behavioral norms in multi-agent LLM systems.…

• When Domains Interact: Asymmetric and Order-Sensitive Cross-Domain Effects in Reinforcement Learning for Reasoning

  Wang Yang, Shouren Wang, Chaoda Song, Chuang Ma, Xinpeng Li, Nengbo Wang, Kaixiong Zhou, Vipin Chaudhary, Xiaotian Han · 3 de febrero de 2026

  Group Relative Policy Optimization (GRPO) has become a key technique for improving reasoning abilities in large language models, yet its behavior under different domain sequencing strategies is poorly understood. In particular, the impact of sequential (one domain at a time) versus mixed-domain (mul…

• Reinforcement Learning for Durable Algorithmic Recourse

  Marina Ceccon, Alessandro Fabris, Goran Radanovi\'c, Asia J. Biega, Gian Antonio Susto · 3 de febrero de 2026

  Algorithmic recourse seeks to provide individuals with actionable recommendations that increase their chances of receiving favorable outcomes from automated decision systems (e.g., loan approvals). While prior research has emphasized robustness to model updates, considerably less attention has been …

• The Multiple Ticket Hypothesis: Random Sparse Subnetworks Suffice for RLVR

  Israel Adewuyi, Solomon Okibe, Vladmir Ivanov · 3 de febrero de 2026

  The Lottery Ticket Hypothesis demonstrated that sparse subnetworks can match full-model performance, suggesting parameter redundancy. Meanwhile, in Reinforcement Learning with Verifiable Rewards (RLVR), recent work has shown that updates concentrate on a sparse subset of parameters, which further le…

• Adaptive Rollout Allocation for Online Reinforcement Learning with Verifiable Rewards

  Hieu Trung Nguyen, Bao Nguyen, Wenao Ma, Yuzhi Zhao, Ruifeng She, Viet Anh Nguyen · 3 de febrero de 2026

  Sampling efficiency is a key bottleneck in reinforcement learning with verifiable rewards. Existing group-based policy optimization methods, such as GRPO, allocate a fixed number of rollouts for all training prompts. This uniform allocation implicitly treats all prompts as equally informative, and c…

• ProcMEM: Learning Reusable Procedural Memory from Experience via Non-Parametric PPO for LLM Agents

  Qirui Mi, Zhijian Ma, Mengyue Yang, Haoxuan Li, Yisen Wang, Haifeng Zhang, Jun Wang · 3 de febrero de 2026

  LLM-driven agents demonstrate strong performance in sequential decision-making but often rely on on-the-fly reasoning, re-deriving solutions even in recurring scenarios. This insufficient experience reuse leads to computational redundancy and execution instability. To bridge this gap, we propose Pro…

• Bandwidth-Efficient Multi-Agent Communication through Information Bottleneck and Vector Quantization

  Ahmad Farooq, Kamran Iqbal · 3 de febrero de 2026

  Multi-agent reinforcement learning systems deployed in real-world robotics applications face severe communication constraints that significantly impact coordination effectiveness. We present a framework that combines information bottleneck theory with vector quantization to enable selective, bandwid…

• FAIRGAME: a Framework for AI Agents Bias Recognition using Game Theory

  Alessio Buscemi, Daniele Proverbio, Alessandro Di Stefano, The Anh Han, German Castignani, Pietro Li\`o · 3 de febrero de 2026

  Letting AI agents interact in multi-agent applications adds a layer of complexity to the interpretability and prediction of AI outcomes, with profound implications for their trustworthy adoption in research and society. Game theory offers powerful models to capture and interpret strategic interactio…