Meta tags:
description=PhD student at National University of Singapore.;
author=Chenrui Tie;
keywords=Chenrui Tie,PhD,Research,National University of Singapore;

Headings (most frequently used words):

for, robotic, assembly, prehensile, and, skills, manual2skill, manuals, vision, language, models, learning, se, chenrui, tie, about, news, selected, publications, research, interests, adaptpnp, integrating, non, adaptive, manipulation, connector, aware, general, from, instruction, via, to, read, acquire, furniture, using, et, seed, efficient, trajectory, level, equivariant, diffusion, policy, leveraging, equivariance, 3d, geometric, shape,

Text of the page (most frequently used words):
and (13), assembly (8), chenrui (8), tie (8), chen (8), #robotic (7), lin (6), #vision (5), ruihai (5), for (5), learning (5), skills (5), shao (5), prehensile (5), manipulation (5), equivariance (4), the (4), international (4), conference (4), dong (4), manuals (4), manual2skill (4), language (4), models (4), from (4), university (4), prof (4), work (4), 2025 (4), with (3), 2026 (3), task (3), leveraging (3), ieee (3), hao (3), tasks (3), representations (3), that (3), acquire (3), robotics (3), junting (3), haonan (3), jinxuan (3), zhu (3), icra (3), wang (3), non (3), national (3), singapore (3), our (3), has (3), been (3), accepted (3), tackle (2), which (2), computer (2), propose (2), diffusion (2), policy (2), data (2), efficiency (2), generalization (2), iclr (2), yue (2), seed (2), level (2), framework (2), vlm (2), furniture (2), science (2), systems (2), rss (2), jingxiang (2), guo (2), shengxiang (2), sun (2), connector (2), aware (2), hierarchical (2), enabling (2), generalizable (2), automation (2), adaptpnp (2), publications (2), phd (2), supervised (2), 2024 (2), research (2), physics (2), into (2), before (2), student (2), school (2), about (2), 铁宸睿 (2), built, prism, last, updated, feb, multi, part, geometrically, invariance, fits, natural, characteristic, narrows, solution, space, cvf, iccv, yan, zhao, yushi, geometric, shape, new, method, certain, symmetry, achieving, better, spatial, chongkai, gao, zeyi, boxuan, efficient, trajectory, equivariant, novel, enables, understand, human, designed, yiwei, liu, read, using, utilizes, extract, structured, graph, yiman, pang, zhouhan, zhong, zhongrui, yanbo, yudi, general, instruction, via, model, empowered, adaptively, selects, schedules, various, zixuan, yuran, xinyi, cao, integrating, adaptive, view, all, selected, starting, news, focuses, robots, specifically, aim, improve, incorporating, priors, such, frameworks, enhance, high, abstractions, worth, mentioning, studying, majored, naturally, inspires, integrate, physical, intuitions, efficiently, second, year, computing, this, received, degree, eecs, peking, undergraduate, study, privileged, like, foundation, robot, interests, open, main, menu, services,

Text of the page (random words):
chenrui tie 铁宸睿 chenrui tie 铁宸睿 about publications services cv open main menu chenrui tie phd student national university of singapore research interests robot manipulation foundation models robotic assembly se 3 equivariance non prehensile manipulation like about i am a second year ph d student in the school of computing national university of singapore supervised by prof lin shao before this i received my b s degree from school of eecs peking university in 2024 in my undergraduate study i am privileged to work with prof hao dong and prof he wang my research focuses on enabling robots to efficiently acquire generalizable manipulation skills specifically i aim to 1 improve data efficiency by incorporating physics priors such as se 3 equivariance into learning frameworks and 2 enhance generalization by leveraging high level abstractions from vision language models and hierarchical task representations it s worth mentioning that before studying computer science i majored in physics which naturally inspires me to integrate physical intuitions into robotic learning systems news 2026 01 our work manual2skill and adaptpnp has been accepted by icra 2026 2025 04 our work manual2skill has been accepted by rss 2025 2025 01 our work et seed has been accepted by iclr 2025 2024 08 starting my phd at national university of singapore supervised by prof lin shao selected publications view all adaptpnp integrating prehensile and non prehensile skills for adaptive robotic manipulation jinxuan zhu chenrui tie xinyi cao yuran wang jingxiang guo zixuan chen haonan chen junting chen ruihai wu lin shao ieee international conference on robotics automation icra a vision language model empowered framework that adaptively selects and schedules prehensile and non prehensile skills for various manipulation tasks manual2skill connector aware general robotic assembly from instruction manuals via vision language models chenrui tie shengxiang sun yudi lin yanbo wang zhongrui li zhouhan zhong jinxuan zhu yiman pang haonan chen junting chen ruihai wu lin shao ieee international conference on robotics automation icra we utilizes a vlm to extract structured connector aware hierarchical graph representations from assembly manuals enabling generalizable robotic assembly manual2skill learning to read manuals and acquire robotic skills for furniture assembly using vision language models chenrui tie shengxiang sun jinxuan zhu yiwei liu jingxiang guo yue hu haonan chen junting chen ruihai wu lin shao robotics science and systems rss we propose a novel framework that enables vlm to understand human designed manuals and acquire robotic skills for furniture assembly tasks et seed efficient trajectory level se 3 equivariant diffusion policy chenrui tie yue chen ruihai wu boxuan dong zeyi li chongkai gao hao dong international conference on learning representations iclr we propose a new diffusion policy method to tackle tasks with certain symmetry achieving better data efficiency and spatial generalization leveraging se 3 equivariance for learning 3d geometric shape assembly ruihai wu chenrui tie yushi du yan zhao hao dong ieee cvf international conference on computer vision iccv we tackle multi part geometrically assembly task leveraging se 3 equivariance and invariance which fits the natural characteristic of the task and narrows the solution space last updated feb 09 2026 built with prism