Elizaveta Kovtun, Matvey Konovalov, Andrey Sakhovskiy et al.cs.RO
Tactile sensing is an essential modality for robots performing contact-rich, dexterous manipulation, particularly under visual occlusion.
Hengyuan Zhang, Chenming Shang, Zunhai Su et al.cs.CL
Best-of-N is a widely used inference strategy for complex reasoning, whose effectiveness depends on whether sampled candidates can cover diverse and high-quality reasoning paths.
Noga Kertes, Daphna Link Sourani, Alex M. Bronstein et al.cs.CV
Diffusion-weighted imaging (DWI) remains highly vulnerable to subject motion, particularly in time-efficient protocols and in motion-prone populations.
Mingke Lu, Anxing Xiao, David Hsucs.RO
Understanding natural human instructions is crucial for deploying robots in human-centric environments. We study multimodal instruction grounding, where language and gesture provide complementary but uncertain cues.
Pawan K. Tripathi, Hemant Sharma, Andrew Chuang et al.cs.AI
Synchrotron data reduction, detector calibration followed by azimuthal integration of terabyte-scale diffraction series, is a multi-step, expert-bound bottleneck that increasingly limits the science rate of user facilities.
Mehedi Khan, Md. Shariful Islam Bhuyancs.CL
A RAG system can hand you the right answer and cite a source it did not actually use. Models output these unfaithful citations via post-rationalization: they write the answer first and then attach a citation to whatever passage looks close enough.
Haoxuan Li, Sixu Yan, Lianghui Zhu et al.cs.RO
Vision-Language-Action (VLA) models have demonstrated remarkable generalization in robotic manipulation via large-scale multimodal pretraining. However, VLA models are mainly trained on 2D-centric observations, which inherently constrains their capacity for precise spatial manipulation.
Bowei Li, Yuner Zhang, Changliu Liucs.RO
Adapting visuomotor policies to new manipulation tasks often requires substantial manual engineering or teleoperated data collection. Simulation can provide task-specific data at scale, but constructing the scene, designing expert behavior, and configuring data generation still require significant per-task effort.
Gehao Zhang, Weikai Huang, Shailesh Shailesh et al.cs.RO
Manipulation policies must know which objects matter and where they are, yet the pretrained backbones that current robot foundation models build on, from language in vision-language-action models (VLAs) to video generation in world-action models (WAMs), do not directly require this metric grounding, leaving it to be l…
Shengcheng Luo, Xiaoyang Cheng, Hong Ying et al.cs.RO
Human demonstrations offer a scalable way to collect manipulation data, but their contacts may be unstable or infeasible when transferred to a robot hand. Collecting demonstrations directly on the target robot avoids this mismatch, but substantially increases the cost of data collection.
Andrew Nguyen, Yonghyeon Lee, Sangbae Kimcs.RO
Dexterous grasping requires deciding where to grasp, reaching the target, and maintaining stable contact. We connect these stages through a compact three-point interface that separates global geometric reasoning from local contact control.
Yuchen Song, Aditya Mittal, Unnat Jaincs.RO
Behavior cloning for robot manipulation relies on expert demonstrations. However, for tasks that require dynamic stability, precise contact timing, or dexterous coordination, human operators may find it hard or even impossible to collect data.
Yanhou Lai, Fucai Zhu, Ruiqiang Wang et al.cs.RO
Visuomotor policies trained from a few demonstrations may reproduce demonstrated trajectories without reliably following changes in object position. Existing approaches with explicit attention typically obtain spatial priors from human annotation or visual models.
Foundation Model, Li Auto Inccs.CV
Physical AI requires models to ground visual and linguistic understanding in real-world environments while accounting for environmental constraints and execution feedback.
Chuyang Xiao, Haotian Zhan, Sriram Krishna et al.cs.RO
Learning robot manipulation policies from human video demonstrations constitutes a promising avenue for scalable robot learning.
Yiran Wang, Xingyilang Yin, Junfu Pu et al.cs.CV
Modern video games provide a measurable testbed for AI models, combining abilities of visual understanding, instruction decomposition, goal planning, and precise action control over multiple temporal horizons.
Huiqiong Li, Zhiting Mei, Anirudha Majumdar et al.cs.RO
Robotic foundation models achieve impressive performance on standard manipulation benchmarks, yet these evaluations typically assume clean, timely, and consistent visual observations throughout execution.
Dorian Benhamou Goldfajn, Mason Nakamura, Saaduddin Mahmud et al.cs.RO
Multi-robot collaboration could enable more efficient and scalable solutions to complex robotic tasks, but collaboration under partial observability remains challenging. Natural-language communication offers a promising approach to coordinating robots under partial observability.
Yixin Zheng, Jiangran Lyu, Yuntian Deng et al.cs.RO
World Action Models (WAMs) jointly generate robot actions and predict future world states, transferring priors from video pretraining to robot control.
Haoran Yuan, Zekai Wang, Boning Shao et al.cs.RO
Dexterous manipulation depends on contact dynamics that are often only partially observable from vision. Recent World-Action Models (WAMs) couple predictive video world modeling with action generation, but remain largely vision-centric and therefore cannot directly model these contact dynamics.
20 papers listed