Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2602.09021 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914401740652544 |
|---|---|
| author | Yu, Checheng Sima, Chonghao Jiang, Gangcheng Zhang, Hai Mai, Haoguang Li, Hongyang Wang, Huijie Chen, Jin Wu, Kaiyang Chen, Li Zhao, Lirui Shi, Modi Luo, Ping Bu, Qingwen Peng, Shijia Li, Tianyu Yuan, Yibo |
| author_facet | Yu, Checheng Sima, Chonghao Jiang, Gangcheng Zhang, Hai Mai, Haoguang Li, Hongyang Wang, Huijie Chen, Jin Wu, Kaiyang Chen, Li Zhao, Lirui Shi, Modi Luo, Ping Bu, Qingwen Peng, Shijia Li, Tianyu Yuan, Yibo |
| contents | High-reliability long-horizon robotic manipulation has traditionally relied on large-scale data and compute to understand complex real-world dynamics. However, we identify that the primary bottleneck to real-world robustness is not resource scale alone, but the distributional shift among the human demonstration distribution, the inductive bias learned by the policy, and the test-time execution distribution -- a systematic inconsistency that causes compounding errors in multi-stage tasks. To mitigate these inconsistencies, we propose $χ_{0}$, a resource-efficient framework with effective modules designated to achieve production-level robustness in robotic manipulation. Our approach builds off three technical pillars: (i) Model Arithmetic, a weight-space merging strategy that efficiently soaks up diverse distributions of different demonstrations, varying from object appearance to state variations; (ii) Stage Advantage, a stage-aware advantage estimator that provides stable, dense progress signals, overcoming the numerical instability of prior non-stage approaches; and (iii) Train-Deploy Alignment, which bridges the distribution gap via spatio-temporal augmentation, heuristic DAgger corrections, and temporal chunk-wise smoothing. $χ_{0}$ enables two sets of dual-arm robots to collaboratively orchestrate long-horizon garment manipulation, spanning tasks from flattening, folding, to hanging different clothes. Our method exhibits high-reliability autonomy; we are able to run the system from arbitrary initial state for consecutive 24 hours non-stop. Experiments validate that $χ_{0}$ surpasses the state-of-the-art $π_{0.5}$ in success rate by nearly 250%, with only 20-hour data and 8 A100 GPUs. Code, data and models will be released to facilitate the community. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_09021 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | $χ_{0}$: Resource-Aware Robust Manipulation via Taming Distributional Inconsistencies Yu, Checheng Sima, Chonghao Jiang, Gangcheng Zhang, Hai Mai, Haoguang Li, Hongyang Wang, Huijie Chen, Jin Wu, Kaiyang Chen, Li Zhao, Lirui Shi, Modi Luo, Ping Bu, Qingwen Peng, Shijia Li, Tianyu Yuan, Yibo Robotics Computer Vision and Pattern Recognition High-reliability long-horizon robotic manipulation has traditionally relied on large-scale data and compute to understand complex real-world dynamics. However, we identify that the primary bottleneck to real-world robustness is not resource scale alone, but the distributional shift among the human demonstration distribution, the inductive bias learned by the policy, and the test-time execution distribution -- a systematic inconsistency that causes compounding errors in multi-stage tasks. To mitigate these inconsistencies, we propose $χ_{0}$, a resource-efficient framework with effective modules designated to achieve production-level robustness in robotic manipulation. Our approach builds off three technical pillars: (i) Model Arithmetic, a weight-space merging strategy that efficiently soaks up diverse distributions of different demonstrations, varying from object appearance to state variations; (ii) Stage Advantage, a stage-aware advantage estimator that provides stable, dense progress signals, overcoming the numerical instability of prior non-stage approaches; and (iii) Train-Deploy Alignment, which bridges the distribution gap via spatio-temporal augmentation, heuristic DAgger corrections, and temporal chunk-wise smoothing. $χ_{0}$ enables two sets of dual-arm robots to collaboratively orchestrate long-horizon garment manipulation, spanning tasks from flattening, folding, to hanging different clothes. Our method exhibits high-reliability autonomy; we are able to run the system from arbitrary initial state for consecutive 24 hours non-stop. Experiments validate that $χ_{0}$ surpasses the state-of-the-art $π_{0.5}$ in success rate by nearly 250%, with only 20-hour data and 8 A100 GPUs. Code, data and models will be released to facilitate the community. |
| title | $χ_{0}$: Resource-Aware Robust Manipulation via Taming Distributional Inconsistencies |
| topic | Robotics Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2602.09021 |