Ego Guidance: Enhancing Robot Policy Learning with Action-Free Human Egocentric Videos
Organizations: Zhejiang University · Ant Group · University of Chinese Academy of Sciences · Nanyang Technology University ·
Abstract
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.
Figures & tables
| Cucumber | Flower | Toaster | Cup | Bottle | Total | |
| Transfer | Handover | Loading | Hanging | Uncapping | ||
| # Demos | 220 | 160 | 150 | 139 | 57 | – |
| 9/10 | 9/10 | 2/10 | 1/10 | 1/10 | 22/50 | |
| Ours | 9/10 | 8/10 | 2/10 | 0/10 | 1/10 | 20/50 |
| Cucumber Transfer | Flower Handover | |||||||
| Method | Default | Illum. | Backgr. | Distr. | Default | Illum. | Backgr. | Distr. |
| 9 | 7 | 7 | 4 | 9 | 5 | 2 | 3 | |
| Ours | 9 | 9 | 8 | 5 | 8 | 2 | 2 | 0 |
Appendix figures & tables2 assets
Supplementary material from the paper’s appendix.
Appendix
| Target | Encoder | Pretraining objective | Res. | ||
|---|---|---|---|---|---|
| Qwen-ViT | Qwen3.5-2B vision encoder [ 12 ] | vision–language (VLM) | 224 | 49 | 2048 |
| SigLIP2 | So400m/16 [ 32 ] | language–image | 256 | 256 | 1152 |
| DINOv2 | ViT-g/14 [ 33 ] | self-distillation | 224 | 257 | 1536 |
| LingBot | LingBot-Vision ViT-g/16 [ 29 ] | masked boundary modeling | 512 | 1024 | 1536 |
| V-JEPA2 | ViT-g/16 [ 26 ] | latent video prediction | 256 | 256 | 1408 |
| Wan-VAE | Wan2.2 VAE [ 28 ] | pixel reconstruction | 256 | 256 | 48 |
| Optimizer | AdamW, , , weight decay |
|---|---|
| Learning rate | LLM ; experts ; other |
| Schedule | 500 warmup steps, cosine decay to |
| Pretraining steps | 20,000 |
| Downstream fine-tuning steps | 20,000 |
| Pretraining global batch size | 1,024 (1:1 robot/egocentric) |
| Gradient clipping | 1.0 |
Robot Policy Learning
Robot policy learning develops controllers that map sensory observations, task goals, and interaction history to actions for embodied robots.
See all 405 papers