Value prediction is an important robotics problem, wherein we can determine how useful a state is to the successful execution of a task in the future. The key insight that Jason and his co-authors showed in this paper was simple:

* Large vision-language models already encode a ton of information about task completion from being trained on a wide range of human image and video data

* We can use these on robotics tasks to estimate progress, which is a useful surrogate for the “value” of any individual state.

Here’s the abstract:

Predicting temporal progress from visual trajectories is important for intelligent robots that can learn, adapt, and improve. However, learning such progress estimator, or temporal value function, across different tasks and domains requires both a large amount of diverse data and methods which can scale and generalize. To address these challenges, we present Generative Value Learning (GVL), a universal value function estimator that leverages the world knowledge embedded in vision-language models (VLMs) to predict task progress. Naively asking a VLM to predict values for a video sequence performs poorly due to the strong temporal correlation between successive frames. Instead, GVL poses value estimation as a temporal ordering problem over shuffled video frames; this seemingly more challenging task encourages VLMs to more fully exploit their underlying semantic and temporal grounding capabilities to differentiate frames based on their perceived task progress, consequently producing significantly better value predictions. Without any robot or task specific training, GVL can in-context zero-shot and few-shot predict effective values for more than 300 distinct real-world tasks across diverse robot platforms, including challenging bimanual manipulation tasks. Furthermore, we demonstrate that GVL permits flexible multi-modal in-context learning via examples from heterogeneous tasks and embodiments, such as human videos. The generality of GVL enables various downstream applications pertinent to visuomotor policy learning, including dataset filtering, success detection, and value-weighted regression -- all without any model training or finetuning.

Watch on YouTube

Listen on Spotify

Original post on Twitter/X

Project Site

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit robopapers.substack.com

Ep #4: Vision Language Models are In-Context Value Learners

RoboPapers

Ep #4: Vision Language Models are In-Context Value Learners