Ego Guidance: Enhancing Robot Policy Learning with Action-Free Human Egocentric Videos
Abstract
A mixture-of-transformers architecture jointly predicts future visual states and actions, leveraging pretrained vision-language representations and egocentric videos to improve robot policy learning without action supervision.
In this paper, we investigate how human egocentric videos can benefit robot policy learning without action supervision. Building on a pretrained VLM, we propose Shared Prediction on Action and Representation (SPAR), an MoT architecture in which future visual state prediction can provide independent and complementary learning signals alongside action prediction. We find that representations learned from egocentric videos improve downstream performance on out-of-distribution robot tasks, with further gains as video data scales. We also study the effects of objective choices and data curation, providing insights into how to leverage action-free videos for scalable robot learning.
Community
Can inexpensive, readily available human egocentric videos benefit embodied tasks without explicit action annotations? We design an architecture to uncover the implicit mutual information between actions and video, and explore its potential for generalization and scaling.
Project Page: https://llong-cs.github.io/EgoGuidance/
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
- WorldBridge: Bridging Video and Action for Task Generalization in World Action Models (2026)
- SWAM: Scaling World Action Models with Cross-Modal Alignment (2026)
- Ego-CEVR: Learning Cross-Embodiment Visual Representations from Human Egocentric Videos (2026)
- FloMo: 3D Scene Flow as a World Action Model Intermediate for Learning from Human Video (2026)
- PDRVLA: A Vision-Language-Action Model with Physical Dynamics Representations (2026)
- TS-WAM: Temporal Latent Actions and World-Action Coupling for Vision-Language Robot Control (2026)
- V-JEPA Policy: Building Effective World-Action Models on Predictive Visual Latents (2026)
Please give a thumbs up to this comment if you found it helpful!
If you want recommendations for any Paper on Hugging Face checkout this Space
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: @librarian-bot recommend
Get this paper in your agent:
hf papers read 2611.13685 Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash Models citing this paper 0
No model linking this paper
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper
Collections including this paper 0
No Collection including this paper