Chelsea Finn

Researcher@Physical Intelligence·美国·OpenReview

研究方向

foundation models for robotics · imitation learning · reinforcement learning · robustness · distribution shift · model-based reinforcement learning · video prediction · meta-learning · learning to learn · one-shot learning · self-supervised learning · unsupervised learning

Chelsea Finn

EXPO: Stable Reinforcement Learning with Expressive Policies

Cosmos Policy: Fine-Tuning Video Models for Visuomotor Control and Planning

Ctrl-World: A Controllable Generative World Model for Robot Manipulation

RLAD: Training LLMs to Discover Abstractions for Solving Reasoning Problems

Polychromic Objectives for Reinforcement Learning

FSPO: Few-Shot Optimization of Synthetic Preferences Effectively Personalizes to Real Users

Value Flows

Scaling up Memory for Robotic Control via Experience Retrieval

Just Enough Thinking: Efficient Reasoning with Adaptive Length Penalties Reinforcement Learning

What Matters for Batch Online Reinforcement Learning in Robotics?

Feedback Descent: Open-Ended Text Optimization via Pairwise Comparison

Inference-Time Alignment via Hypothesis Reweighting

Reinforcement Learning via Implicit Imitation Guidance

Posterior Behavioral Cloning: Pretraining BC Policies for Efficient RL Finetuning

RoboReward: A Dataset and Benchmark for Vision-Language Reward Models in Robotics

SteerVLA: Steering Vision-Language-Action Models Toward Effective Long-Tail Driving

MJ-Video: Benchmarking and Rewarding Video Generation with Fine-Grained Video Preference

Bidirectional Decoding: Improving Action Chunking via Guided Test-Time Sampling

Latent Diffusion Planning for Imitation Learning

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

Grounding by Trying: LLMs with Reinforcement Learning-Enhanced Retrieval

Range, not Independence, Drives Modularity in Biologically Inspired Representations

MJ-Bench: Is Your Multimodal Reward Model Really a Good Judge for Text-to-Image Generation?

Agent Q: Advanced Reasoning and Learning for Autonomous AI Agents

Test-Time Alignment via Hypothesis Reweighting

Improving the Efficiency of Test-Time Search in LLMs with Backtracking

Parameterization Agnostic RL

Generative Reward Models

Latent Diffusion Planning for Imitation Learning