Ego2Robot: Scalable Robot Data Synthesis from Egocentric Human Data
- 类型:arxiv
- 标识:2608.02580
- 链接:https://arxiv.org/abs/2608.02580
- 主分类:engineering
- 形态:method
- 被引:0
- 被引来源:Semantic Scholar
- S2被引:0
- 影响力被引:0
- TLDR:Experiments show that joint pretraining on Ego2Robot-synthesized and robot data consistently improves out-of-distribution generalization across multiple perturbation types, with benefits validated on real-robot deployment.
- 待LLM分类:否
- 标题中文:Ego2Robot:从第一人称人类数据出发的可扩展机器人数据合成
- TLDR中文:实验表明,在 Ego2Robot 合成数据与机器人数据上的联合预训练在多种扰动类型上一致提升分布外泛化能力,并在真实机器人部署中得到验证。
- 来源文件:
- /inbox/tom/_candidates/2026-08-06-agent-rag-longcontext-candidates.json
- [S2 enrich]