About

Li Ke (柯立) is an Assistant Professor in the Elementary Teacher Education Program (ELTEP) at the University of Washington College of Education, with a focus on Science and Mathematics Teacher Education and Research. His research centers around promoting meaningful science teaching and learning through disciplinary practices such as modeling, especially within the context of complex societal issues like pandemics and climate change. Dr. Ke explores these themes through three interconnected dimensions: curriculum development, student learning, and teacher professional development, emphasizing the active role of both students and teachers as epistemic agents in constructing knowledge to understand and address societal challenges. His work includes leading projects that integrate data science and issue-based learning into K-12 science education, aiming to prepare informed and responsible citizens. Notably, he is involved in a National Science Foundation-funded project that partners with educators and disciplinary experts to co-design data-driven units on wildfires, and another project investigating how high school students learn about viral epidemics through scientific modeling. Recognized for his research contributions, Dr. Ke received the NARST’s Early Career Research Award in 2026. His academic background includes a Ph.D. from Michigan State University, a Master’s from the University of Wisconsin-Madison, and a Bachelor’s in Chemistry from Fudan University.

Research topics

• Artificial Intelligence
• Computer Science
• Machine Learning
• Mathematics
• Philosophy
• Human–computer interaction
• Psychology
• Theoretical computer science
• Simulation
• Physics
• Algorithm

Selected publications

• RL Token: Bootstrapping Online RL with Vision-Language-Action Models

  arXiv (Cornell University) · 2026-04-24

  preprintOpen accessSenior author

  Vision-language-action (VLA) models can learn to perform diverse manipulation skills "out of the box," but achieving the precision and speed that real-world tasks demand requires further fine-tuning -- for example, via reinforcement learning (RL). We introduce a lightweight method that enables sample-efficient online RL fine-tuning of pretrained VLAs using just a few hours of real-world practice. We (1) adapt the VLA to expose an "RL token," a compact readout representation that preserves task-relevant pretrained knowledge while serving as an efficient interface for online RL, and (2) train a small actor-critic head on this RL token to refine the actions, while anchoring the learned policy to the VLA. Online RL with the RL token (RLT) makes it possible to fine-tune even large VLAs with RL quickly and efficiently. Across four real-robot tasks (screw installation, zip tie fastening, charger insertion, and Ethernet insertion), RLT improves the speed on the hardest part of the task by up to 3x and raises success rates significantly within minutes to a few hours of practice. It can even surpass the speed of human teleoperation on some of the tasks.

  PublisherDOI

• RL Token: Bootstrapping Online RL with Vision-Language-Action Models

  arXiv (Cornell University) · 2026-04-24

  articleOpen accessSenior author

  Vision-language-action (VLA) models can learn to perform diverse manipulation skills "out of the box," but achieving the precision and speed that real-world tasks demand requires further fine-tuning -- for example, via reinforcement learning (RL). We introduce a lightweight method that enables sample-efficient online RL fine-tuning of pretrained VLAs using just a few hours of real-world practice. We (1) adapt the VLA to expose an "RL token," a compact readout representation that preserves task-relevant pretrained knowledge while serving as an efficient interface for online RL, and (2) train a small actor-critic head on this RL token to refine the actions, while anchoring the learned policy to the VLA. Online RL with the RL token (RLT) makes it possible to fine-tune even large VLAs with RL quickly and efficiently. Across four real-robot tasks (screw installation, zip tie fastening, charger insertion, and Ethernet insertion), RLT improves the speed on the hardest part of the task by up to 3x and raises success rates significantly within minutes to a few hours of practice. It can even surpass the speed of human teleoperation on some of the tasks.

  PublisherOA PDF

• $π_{0.7}$: a Steerable Generalist Robotic Foundation Model with Emergent Capabilities

  ArXiv.org · 2026-04-16

  articleOpen access

  We present a new robotic foundation model, called $π_{0.7}$, that can enable strong out-of-the-box performance in a wide range of scenarios. $π_{0.7}$ can follow diverse language instructions in unseen environments, including multi-stage tasks with various kitchen appliances, provide zero-shot cross-embodiment generalization, for example enabling a robot to fold laundry without seeing the task before, and perform challenging tasks such as operating an espresso machine out of the box at a level of performance that matches much more specialized RL-finetuned models. The main idea behind $π_{0.7}$ is to use diverse context conditioning during training. This conditioning information, contained in the prompt, makes it possible to steer the model precisely to perform many tasks with different strategies. It is conditioned not just on a language command that describes what it should do, but on additional multimodal information that also describes the manner or strategy in which it should do it, including metadata about task performance and subgoal images. This enables $π_{0.7}$ to use very diverse data, including demonstrations, potentially suboptimal (autonomous) data including failures, and data from non-robot sources. Our experiments evaluate $π_{0.7}$ across numerous tasks with multiple robot platforms, on tasks that require speed and dexterity, language following, and compositional task generalization.

  PublisherOA PDF

• $π_{0.7}$: a Steerable Generalist Robotic Foundation Model with Emergent Capabilities

  arXiv (Cornell University) · 2026-04-16

  preprintOpen access

  We present a new robotic foundation model, called $π_{0.7}$, that can enable strong out-of-the-box performance in a wide range of scenarios. $π_{0.7}$ can follow diverse language instructions in unseen environments, including multi-stage tasks with various kitchen appliances, provide zero-shot cross-embodiment generalization, for example enabling a robot to fold laundry without seeing the task before, and perform challenging tasks such as operating an espresso machine out of the box at a level of performance that matches much more specialized RL-finetuned models. The main idea behind $π_{0.7}$ is to use diverse context conditioning during training. This conditioning information, contained in the prompt, makes it possible to steer the model precisely to perform many tasks with different strategies. It is conditioned not just on a language command that describes what it should do, but on additional multimodal information that also describes the manner or strategy in which it should do it, including metadata about task performance and subgoal images. This enables $π_{0.7}$ to use very diverse data, including demonstrations, potentially suboptimal (autonomous) data including failures, and data from non-robot sources. Our experiments evaluate $π_{0.7}$ across numerous tasks with multiple robot platforms, on tasks that require speed and dexterity, language following, and compositional task generalization.

  PublisherDOI

• ATK: Automatic Task-driven Keypoint Selection for Robust Policy Learning

  ArXiv.org · 2025-06-16

  preprintOpen access

  Visuomotor policies often suffer from perceptual challenges, where visual differences between training and evaluation environments degrade policy performance. Policies relying on state estimations, like 6D pose, require task-specific tracking and are difficult to scale, while raw sensor-based policies may lack robustness to small visual disturbances. In this work, we leverage 2D keypoints--spatially consistent features in the image frame--as a flexible state representation for robust policy learning and apply it to both sim-to-real transfer and real-world imitation learning. However, the choice of which keypoints to use can vary across objects and tasks. We propose a novel method, ATK, to automatically select keypoints in a task-driven manner so that the chosen keypoints are predictive of optimal behavior for the given task. Our proposal optimizes for a minimal set of keypoints that focus on task-relevant parts while preserving policy performance and robustness. We distill expert data (either from an expert policy in simulation or a human expert) into a policy that operates on RGB images while tracking the selected keypoints. By leveraging pre-trained visual modules, our system effectively encodes states and transfers policies to the real-world evaluation scenario despite wide scene variations and perceptual challenges such as transparent objects, fine-grained tasks, and deformable objects manipulation. We validate ATK on various robotic tasks, demonstrating that these minimal keypoint representations significantly improve robustness to visual disturbances and environmental variations. See all experiments and more details at https://yunchuzhang.github.io/ATK/.

  PublisherOA PDFDOI

• $π_{0.5}$: a Vision-Language-Action Model with Open-World Generalization

  ArXiv.org · 2025-04-22 · 2 citations

  preprintOpen access

  In order for robots to be useful, they must perform practically relevant tasks in the real world, outside of the lab. While vision-language-action (VLA) models have demonstrated impressive results for end-to-end robot control, it remains an open question how far such models can generalize in the wild. We describe $π_{0.5}$, a new model based on $π_{0}$ that uses co-training on heterogeneous tasks to enable broad generalization. $π_{0.5}$\ uses data from multiple robots, high-level semantic prediction, web data, and other sources to enable broadly generalizable real-world robotic manipulation. Our system uses a combination of co-training and hybrid multi-modal examples that combine image observations, language commands, object detections, semantic subtask prediction, and low-level actions. Our experiments show that this kind of knowledge transfer is essential for effective generalization, and we demonstrate for the first time that an end-to-end learning-enabled robotic system can perform long-horizon and dexterous manipulation skills, such as cleaning a kitchen or bedroom, in entirely new homes.

  PublisherOA PDFDOI

• Hi Robot: Open-Ended Instruction Following with Hierarchical Vision-Language-Action Models

  ArXiv.org · 2025-02-26

  preprintOpen access

  Generalist robots that can perform a range of different tasks in open-world settings must be able to not only reason about the steps needed to accomplish their goals, but also process complex instructions, prompts, and even feedback during task execution. Intricate instructions (e.g., "Could you make me a vegetarian sandwich?" or "I don't like that one") require not just the ability to physically perform the individual steps, but the ability to situate complex commands and feedback in the physical world. In this work, we describe a system that uses vision-language models in a hierarchical structure, first reasoning over complex prompts and user feedback to deduce the most appropriate next step to fulfill the task, and then performing that step with low-level actions. In contrast to direct instruction following methods that can fulfill simple commands ("pick up the cup"), our system can reason through complex prompts and incorporate situated feedback during task execution ("that's not trash"). We evaluate our system across three robotic platforms, including single-arm, dual-arm, and dual-arm mobile robots, demonstrating its ability to handle tasks such as cleaning messy tables, making sandwiches, and grocery shopping. Videos are available at https://www.pi.website/research/hirobot

  PublisherOA PDFDOI

• $π^{*}_{0.6}$: a VLA That Learns From Experience

  ArXiv.org · 2025-11-18

  preprintOpen access

  We study how vision-language-action (VLA) models can improve through real-world deployments via reinforcement learning (RL). We present a general-purpose method, RL with Experience and Corrections via Advantage-conditioned Policies (RECAP), that provides for RL training of VLAs via advantage conditioning. Our method incorporates heterogeneous data into the self-improvement process, including demonstrations, data from on-policy collection, and expert teleoperated interventions provided during autonomous execution. RECAP starts by pre-training a generalist VLA with offline RL, which we call $π^{*}_{0.6}$, that can then be specialized to attain high performance on downstream tasks through on-robot data collection. We show that the $π^{*}_{0.6}$ model trained with the full RECAP method can fold laundry in real homes, reliably assemble boxes, and make espresso drinks using a professional espresso machine. On some of the hardest tasks, RECAP more than doubles task throughput and roughly halves the task failure rate.

  PublisherOA PDFDOI

• π₀: A Vision-Language-Action Flow Model for General Robot Control

  2025-06-21 · 124 citations

  articleOpen access
  PublisherDOI

• Overcoming the Sim-to-Real Gap: Leveraging Simulation to Learn to Explore for Real-World RL

  arXiv (Cornell University) · 2024-10-26

  preprintOpen access

  In order to mitigate the sample complexity of real-world reinforcement learning, common practice is to first train a policy in a simulator where samples are cheap, and then deploy this policy in the real world, with the hope that it generalizes effectively. Such \emph{direct sim2real} transfer is not guaranteed to succeed, however, and in cases where it fails, it is unclear how to best utilize the simulator. In this work, we show that in many regimes, while direct sim2real transfer may fail, we can utilize the simulator to learn a set of \emph{exploratory} policies which enable efficient exploration in the real world. In particular, in the setting of low-rank MDPs, we show that coupling these exploratory policies with simple, practical approaches -- least-squares regression oracles and naive randomized exploration -- yields a polynomial sample complexity in the real world, an exponential improvement over direct sim2real transfer, or learning without access to a simulator. To the best of our knowledge, this is the first evidence that simulation transfer yields a provable gain in reinforcement learning in settings where direct sim2real transfer fails. We validate our theoretical results on several realistic robotic simulators and a real-world robotic sim2real task, demonstrating that transferring exploratory policies can yield substantial gains in practice as well.

  PublisherOA PDFDOI

Frequent coauthors

• Siddhartha S Srinivasa

  22 shared

• A. Deshpande

  8 shared

• Sanjiban Choudhury

  5 shared

• Gilwoo Lee

  University of Washington

  5 shared

• Matt Barnes

  University of Washington

  5 shared

• Yunchu Zhang

  4 shared

• Tapomayukh Bhattacharjee

  4 shared

• Abhishek Gupta

  4 shared

Awards & honors

• NARST’s Early Career Research Award (2026)