REVIEW 3 cited by
Grounding Video Models to Actions through Goal Conditioned Exploration
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Large video models, pretrained on massive amounts of Internet video, provide a rich source of physical knowledge about the dynamics and motions of objects and tasks. However, video models are not grounded in the embodiment of an agent, and do not describe how to actuate the world to reach the visual states depicted in a video. To tackle this problem, current methods use a separate vision-based inverse dynamic model trained on embodiment-specific data to map image states to actions. Gathering data to train such a model is often expensive and challenging, and this model is limited to visual settings similar to the ones in which data are available. In this paper, we investigate how to directly ground video models to continuous actions through self-exploration in the embodied environment -- using generated video states as visual goals for exploration. We propose a framework that uses trajectory level action generation in combination with video guidance to enable an agent to solve complex tasks without any external supervision, e.g., rewards, action labels, or segmentation masks. We validate the proposed approach on 8 tasks in Libero, 6 tasks in MetaWorld, 4 tasks in Calvin, and 12 tasks in iThor Visual Navigation. We show how our approach is on par with or even surpasses multiple behavior cloning baselines trained on expert demonstrations while without requiring any action annotations.
Forward citations
Cited by 3 Pith papers
-
Generative Visual Foresight Meets Task-Agnostic Pose Estimation in Robotic Table-Top Manipulation
GVF-TAPE predicts future RGB-D frames from an image and text, then extracts end-effector poses to control a robot, achieving strong success rates without action-labeled data.
-
Medical World Model: Generative Simulation of Tumor Evolution for Treatment Planning
MeWM combines a GPT-style policy, a diffusion tumor dynamics model, and a survival analysis heuristic to simulate post-treatment tumor appearance and select TACE treatment plans, improving physician F1-score by 13 points.
-
From World Models to World Action Models: A Concise Tutorial for Robotics
World models are action-conditioned predictors of task-relevant futures; world action models couple those futures to robot actions via four paradigms: imagine-then-execute, feature-conditioned, joint, and auxiliary pr...
Discussion (0). Continue with ORCID to comment.