Chelsea Finn

Stanford University

収録論文 176本 ・ フィジカルAI/ロボット学習

※arXiv著者名で収集。同姓同名の別人の論文が含まれる場合があります。

論文

• 世界モデルを用いたQ学習
  強化学習2026/8/17

  Q学習に世界モデルを組み合わせ、実データのみで学習しつつ想像上の軌道で探索を行うことで、サンプル効率と性能を向上させる手法を提案した。

• シミュレーションでの視覚的器用さの事前学習
  器用操作/事前学習2026/8/16

  VRヘッドセットを使い、人間がシミュレーション内で物体を操作して収集したデータで、多指ロボットハンドの操作ポリシーを事前学習する手法を提案。実機での微調整により、ゼロから学習するよりも高い性能を達成した。

• SpeedTuning: 軽量強化学習によるポリシー実行の高速化
  操作/強化学習2026/8/10

  模倣学習で得たロボット操作ポリシーの実行速度を、追加データ収集なしで強化学習により最適化するフレームワークを提案し、成功率を保ちつつ2.4倍以上の高速化を実証した。

• SpeedTuning: 軽量強化学習によるポリシー実行の高速化
  操作/強化学習2026/8/10

  模倣学習で得たロボット操作ポリシーの実行速度を、追加データ収集なしで強化学習により最適化するフレームワークを提案し、成功率を保ちつつ2.4倍以上の高速化を実証した。

• SpeedTuning: Speeding Up Policy Execution with Lightweight Reinforcement Learning
  2026/8/1
• LLM-as-a-Verifier: A General-Purpose Verification Framework
  2026/7/1
• Improving Robotic Generalist Policies via Flow Reversal Steering
  2026/6/1
• Freeform Preference Learning for Robotic Manipulation
  2026/6/1
• Ego-Pi: VLA Fine-Tuning for Ego-Centric Human and Robot Data
  2026/6/1
• DIRECT: When and Where Should You Allocate Test-Time Compute in Embodied Planners?
  2026/6/1
• CHORUS: Decentralized Multi-Embodiment Collaboration with One VLA Policy
  2026/6/1
• EXPO-FT: Sample-Efficient Reinforcement Learning Finetuning for Vision-Language-Action Models
  2026/5/1
• Open-H-Embodiment: A Large-Scale Dataset for Enabling Foundation Models in Medical Robotics
  2026/4/1
• World Action Verifier: Self-Improving World Models via Forward-Inverse Asymmetry
  2026/4/1
• ${\pi}_{0.7}$: a Steerable Generalist Robotic Foundation Model with Emergent Capabilities
  2026/4/1
• RoboMME: Benchmarking and Understanding Memory for Robotic Generalist Policies
  2026/3/1
• MEM: Multi-Scale Embodied Memory for Vision Language Action Models
  2026/3/1
• Data Analogies Enable Efficient Cross-Embodiment Transfer
  2026/3/1
• Scaling Verification Can Be More Effective than Scaling Policy Learning for Vision-Language-Action Alignment
  2026/2/1
• VLAW: Iterative Co-Improvement of Vision-Language-Action Policy and World Model
  2026/2/1
• SteerVLA: Steering Vision-Language-Action Models in Long-Tail Driving Scenarios
  2026/2/1
• RoboReward: General-Purpose Vision-Language Reward Models for Robotics
  2026/1/1
• Cosmos Policy: Fine-Tuning Video Models for Visuomotor Control and Planning
  2026/1/1
• Posterior Behavioral Cloning: Pretraining BC Policies for Efficient RL Finetuning
  2025/12/1
• Emergence of Human to Robot Transfer in Vision-Language-Action Models
  2025/12/1
• PolaRiS: Scalable Real-to-Sim Evaluations for Generalist Robot Policies
  2025/12/1
• Invariance Co-training for Robot Visual Generalization
  2025/12/1
• $\pi^{*}_{0.6}$: a VLA That Learns From Experience
  2025/11/1
• Ctrl-World: A Controllable Generative World Model for Robot Manipulation
  2025/10/1
• MemER: Scaling Up Memory for Robot Control via Experience Retrieval
  2025/10/1
• SutureBot: A Precision Framework & Benchmark For Autonomous End-to-End Suturing
  2025/10/1
• Self-Guided Action Diffusion
  2025/8/1
• RoboArena: Distributed Real-World Evaluation of Generalist Robot Policies
  2025/6/1
• SRT-H: A Hierarchical Framework for Autonomous Surgery via Language Conditioned Imitation Learning
  2025/5/1
• Learning Long-Context Diffusion Policies via Past-Token Prediction
  2025/5/1
• What Matters for Batch Online Reinforcement Learning in Robotics?
  2025/5/1
• $\pi_{0.5}$: a Vision-Language-Action Model with Open-World Generalization
  2025/4/1
• Latent Diffusion Planning for Imitation Learning
  2025/4/1
• Curating Demonstrations using Online Experience
  2025/3/1
• CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models
  2025/3/1
• Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success
  2025/2/1
• Hi Robot: Open-Ended Instruction Following with Hierarchical Vision-Language-Action Models
  2025/2/1
• FAST: Efficient Action Tokenization for Vision-Language-Action Models
  2025/1/1
• RoboCrowd: Scaling Robot Data Collection through Crowdsourcing
  2024/11/1
• Helpful DoggyBot: Open-World Object Fetching using Legged Robots and Vision-Language Models
  2024/10/1
• ALOHA Unleashed: A Simple Recipe for Robot Dexterity
  2024/10/1
• $\pi_0$: A Vision-Language-Action Flow Model for General Robot Control
  2024/10/1
• D5RL: Diverse Datasets for Data-Driven Deep Reinforcement Learning
  2024/8/1
• Bidirectional Decoding: Improving Action Chunking via Guided Test-Time Sampling
  2024/8/1
• Affordance-Guided Reinforcement Learning via Visual Prompting
  2024/7/1
• Commonsense Reasoning for Legged Robot Adaptation with Vision-Language Models
  2024/7/1
• Mobility VLA: Multimodal Instruction Navigation with Long-Context VLMs and Topological Graphs
  2024/7/1
• Surgical Robot Transformer (SRT): Imitation Learning for Surgical Tasks
  2024/7/1
• Robotic Control via Embodied Chain-of-Thought Reasoning
  2024/7/1
• HumanPlus: Humanoid Shadowing and Imitation from Humans
  2024/6/1
• OpenVLA: An Open-Source Vision-Language-Action Model
  2024/6/1
• To Err is Robotic: Rapid Value-Based Trial-and-Error during Deployment
  2024/6/1
• Octo: An Open-Source Generalist Robot Policy
  2024/5/1
• Evaluating Real-World Robot Manipulation Policies in Simulation
  2024/5/1
• ALOHA 2: An Enhanced Low-Cost Hardware for Bimanual Teleoperation
  2024/5/1
• Efficient Data Collection for Robotic Manipulation via Compositional Generalization
  2024/3/1
• Yell At Your Robot: Improving On-the-Fly from Language Corrections
  2024/3/1
• DROID: A Large-Scale In-The-Wild Robot Manipulation Dataset
  2024/3/1
• Pushing the Limits of Cross-Embodiment Learning for Manipulation and Navigation
  2024/2/1
• PIVOT: Iterative Visual Prompting Elicits Actionable Knowledge for VLMs
  2024/2/1
• MOTO: Offline Pre-training to Online Fine-tuning for Model-based Robot Learning
  2024/1/1
• Mobile ALOHA: Learning Bimanual Mobile Manipulation with Low-Cost Whole-Body Teleoperation
  2024/1/1
• SERL: A Software Suite for Sample-Efficient Robotic Reinforcement Learning
  2024/1/1
• AutoRT: Embodied Foundation Models for Large Scale Orchestration of Robotic Agents
  2024/1/1
• What Makes Pre-Trained Visual Representations Successful for Robust Manipulation?
  2023/12/1
• Adapt On-the-Go: Behavior Modulation for Single-Life Robot Deployment
  2023/11/1
• RT-Trajectory: Robotic Task Generalization via Hindsight Trajectory Sketches
  2023/11/1
• Robot Fine-Tuning Made Easy: Pre-Training Rewards and Policies for Autonomous Real-World Reinforcement Learning
  2023/10/1
• Open X-Embodiment: Robotic Learning Datasets and RT-X Models
  2023/10/1
• Zero-Shot Robotic Manipulation with Pretrained Image-Editing Diffusion Models
  2023/10/1
• RoboCLIP: One Demonstration is Enough to Learn Robot Policies
  2023/10/1
• Offline Retraining for Online RL: Decoupled Policy Learning to Mitigate Exploration Bias
  2023/10/1
• Robot Parkour Learning
  2023/9/1
• Q-Transformer: Scalable Offline Reinforcement Learning via Autoregressive Q-Functions
  2023/9/1
• BridgeData V2: A Dataset for Robot Learning at Scale
  2023/8/1
• Contrastive Example-Based Control
  2023/7/1
• RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control
  2023/7/1
• Giving Robots a Hand: Learning Generalizable Manipulation with Eye-in-Hand Human Video Demonstrations
  2023/7/1
• Waypoint-Based Imitation Learning for Robotic Manipulation
  2023/7/1
• Polybot: Training One Policy Across Robots While Embracing Variability
  2023/7/1
• Decomposing the Generalization Gap in Imitation Learning for Visual Robotic Manipulation
  2023/7/1
• Train Offline, Test Online: A Real Robot Learning Benchmark
  2023/6/1
• Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware
  2023/4/1
• Behavior Retrieval: Few-Shot Imitation Learning by Querying Unlabeled Datasets
  2023/4/1
• A Control-Centric Benchmark for Video Prediction
  2023/4/1
• Open-World Object Manipulation using Pre-trained Vision-Language Models
  2023/3/1
• Self-Improving Robots: End-to-End Autonomous Visuomotor Reinforcement Learning
  2023/3/1
• Language-Driven Representation Learning for Robotics
  2023/2/1
• NeRF in the Palm of Your Hand: Corrective Augmentation for Robotics via Novel-View Synthesis
  2023/1/1
• RT-1: Robotics Transformer for Real-World Control at Scale
  2022/12/1
• Pre-Training for Robots: Offline RL Enables Learning New Tasks from a Handful of Trials
  2022/10/1
• Offline Reinforcement Learning at Multiple Frequencies
  2022/7/1
• A State-Distribution Matching Approach to Non-Episodic Reinforcement Learning
  2022/5/1
• Play it by Ear: Learning Skills amidst Occlusion through Audio-Visual Imitation Learning
  2022/5/1
• Training and Evaluation of Deep Policies using Reinforcement Learning and Generative Models
  2022/4/1
• Do As I Can, Not As I Say: Grounding Language in Robotic Affordances
  2022/4/1
• R3M: A Universal Visual Representation for Robot Manipulation
  2022/3/1
• Vision-Based Manipulators Need to Also See from Their Hands
  2022/3/1
• Policy Architectures for Compositional Generalization in Control
  2022/3/1
• BC-Z: Zero-Shot Task Generalization with Robotic Imitation Learning
  2022/2/1
• Robust Policy Learning over Multiple Uncertainty Sets
  2022/2/1
• How to Leverage Unlabeled Data in Offline Reinforcement Learning
  2022/2/1
• CoMPS: Continual Meta Policy Search
  2021/12/1
• Autonomous Reinforcement Learning: Formalism and Benchmarking
  2021/12/1
• MESA: Offline Meta-RL for Safe Adaptation and Fault Tolerance
  2021/12/1
• Bridge Data: Boosting Generalization of Robotic Skills with Cross-Domain Datasets
  2021/9/1
• Learning Language-Conditioned Robot Behavior from Offline Data and Crowd-Sourced Annotation
  2021/9/1
• Example-Driven Model-Based Reinforcement Learning for Solving Long-Horizon Visuomotor Tasks
  2021/9/1
• Lifelong Robotic Reinforcement Learning by Retaining Experiences
  2021/9/1
• Conservative Data Sharing for Multi-Task Offline Reinforcement Learning
  2021/9/1
• Autonomous Reinforcement Learning via Subgoal Curricula
  2021/7/1
• Visual Adversarial Imitation Learning using Variational Models
  2021/7/1
• Actionable Models: Unsupervised Offline Reinforcement Learning of Robotic Skills
  2021/4/1
• MT-Opt: Continuous Multi-Task Robotic Reinforcement Learning at Scale
  2021/4/1
• Bayesian Meta-Learning for Few-Shot Policy Adaptation Across Robotic Platforms
  2021/3/1
• Greedy Hierarchical Variational Autoencoders for Large-Scale Video Prediction
  2021/3/1
• Learning Generalizable Robotic Reward Functions from "In-The-Wild" Human Videos
  2021/3/1
• COMBO: Conservative Offline Model-Based Policy Optimization
  2021/2/1
• How to Train Your Robot with Deep Reinforcement Learning; Lessons We've Learned
  2021/2/1
• Offline Reinforcement Learning from Images with Latent Space Models
  2020/12/1
• Model-Based Visual Planning with Self-Supervised Functional Distances
  2020/12/1
• Learning Latent Representations to Influence Multi-Agent Interaction
  2020/11/1
• Reinforcement Learning with Videos: Combining Offline Observations with Interaction
  2020/11/1
• Batch Exploration with Examples for Scalable Robotic Reinforcement Learning
  2020/10/1
• Measuring and Harnessing Transference in Multi-Task Learning
  2020/10/1
• MELD: Meta-Reinforcement Learning from Images via Latent State Models
  2020/10/1
• Recovery RL: Safe Reinforcement Learning with Learned Recovery Zones
  2020/10/1
• Learning to be Safe: Deep RL with a Safety Critic
  2020/10/1
• Goal-Aware Prediction: Learning to Model What Matters
  2020/7/1
• Long-Horizon Visual Planning with Goal-Conditioned Hierarchical Predictors
  2020/6/1
• Deep Reinforcement Learning amidst Lifelong Non-Stationarity
  2020/6/1
• Weakly-Supervised Reinforcement Learning for Controllable Behavior
  2020/4/1
• Never Stop Learning: The Effectiveness of Fine-Tuning in Robotic Reinforcement Learning
  2020/4/1
• OmniTact: A Multi-Directional High Resolution Touch Sensor
  2020/3/1
• Rapidly Adaptable Legged Robots via Evolutionary Meta-Learning
  2020/3/1
• Scalable Multi-Task Imitation Learning with Autonomous Improvement
  2020/3/1
• Gradient Surgery for Multi-Task Learning
  2020/1/1
• Learning Predictive Models From Observation and Interaction
  2019/12/1
• Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learning
  2019/10/1
• RoboNet: Large-Scale Multi-Robot Learning
  2019/10/1
• Hierarchical Foresight: Self-Supervised Learning of Long-Horizon Tasks via Visual Subgoal Generation
  2019/9/1
• Guided Meta-Policy Search
  2019/4/1
• End-to-End Robotic Reinforcement Learning without Reward Engineering
  2019/4/1
• Improvisation through Physical Understanding: Using Novel Objects as Tools with Visual Foresight
  2019/4/1
• NoRML: No-Reward Meta Learning
  2019/3/1
• Manipulation by Feel: Touch-Based Control with Deep Predictive Models
  2019/3/1
• Unsupervised Visuomotor Control through Distributional Planning Networks
  2019/2/1
• Deep Online Learning via Meta-Learning: Continual Adaptation for Model-Based RL
  2018/12/1
• Visual Foresight: Model-Based Deep Reinforcement Learning for Vision-Based Robotic Control
  2018/12/1
• Reasoning About Physical Interactions with Object-Oriented Prediction and Planning
  2018/12/1
• Few-Shot Goal Inference for Visuomotor Learning and Planning
  2018/10/1
• Time Reversal as Self-Supervision
  2018/10/1
• One-Shot Hierarchical Imitation Learning of Compound Visuomotor Tasks
  2018/10/1
• Robustness via Retrying: Closed-Loop Robotic Manipulation with Self-Supervised Learning
  2018/10/1
• Universal Planning Networks
  2018/4/1
• Stochastic Adversarial Video Prediction
  2018/4/1
• Learning to Adapt in Dynamic, Real-World Environments Through Meta-Reinforcement Learning
  2018/3/1
• One-Shot Imitation from Observing Humans via Domain-Adaptive Meta-Learning
  2018/2/1
• Deep Reinforcement Learning for Vision-Based Robotic Grasping: A Simulated Comparative Evaluation of Off-Policy Methods
  2018/2/1
• Stochastic Variational Video Prediction
  2017/10/1
• Self-Supervised Visual Planning with Temporal Skip Connections
  2017/10/1
• One-Shot Visual Imitation Learning via Meta-Learning
  2017/9/1
• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
  2017/3/1
• Generalizing Skills with Semi-Supervised Reinforcement Learning
  2016/12/1
• Deep Visual Foresight for Planning Robot Motion
  2016/10/1
• Reset-Free Guided Policy Search: Efficient Deep Reinforcement Learning with Stochastic Initial States
  2016/10/1
• Unsupervised Learning for Physical Interaction through Video Prediction
  2016/5/1
• Guided Cost Learning: Deep Inverse Optimal Control via Policy Optimization
  2016/3/1
• Deep Spatial Autoencoders for Visuomotor Learning
  2015/9/1
• Learning Deep Neural Network Policies with Continuous Memory States
  2015/7/1
• End-to-End Training of Deep Visuomotor Policies
  2015/4/1