12 papers
One Model, Multiple Compute Budgets: Multi-Ratio ConceptMoE for Flexible Compute Across Turns
Yiyang Fu, Zihao Huang, Dongchen Han +4
Large language models typically support different inference budgets through separate model sizes or fixed compute configurations. In multi-turn interactions, however, compute deman…
Rethinking World-Action Model for Compositional and In-Context Robotic Manipulation
Shukai Gong, Xuanran Zhai, Yintianrun Zhang +16
Long-horizon compositional manipulation has become increasingly important for real-world robot deployment, where a single task involves multiple coordinated subtasks. Existing worl…
Multimodal Unlearning Across Vision, Language, Video, and Audio: Survey of Methods, Datasets, and Benchmarks
Nobin Sarwar, Shubhashis Roy Dipta, Zheyuan Liu +1
With the growing adoption of VLMs, DMs, LLMs, and AFMs, these multimodal foundation models can inadvertently encode sensitive, copyrighted, biased, or unsafe cross-modal associatio…
PhysisForcing: Physics Reinforced World Simulator for Robotic Manipulation
Peiwen Zhang, Yufan Deng, Shangkun Sun +11
Video generation models have emerged as a promising paradigm for embodied world simulation. However, both general-domain video generators and robot-specific data fine-tuned models…
HumanScale: Egocentric Human Video Can Outperform Real-Robot Data for Embodied Pretraining
Juncheng Ma, Jianxin Bi, Yufan Deng +19
Embodied foundation models are expected to benefit from data scaling like large language models, but face a much tighter data bottleneck. Teleoperated real-robot trajectories remai…
Fast-dDrive: Efficient Block-Diffusion VLM for Autonomous Driving
Kewei Zhang, Jin Wang, Sensen Gao +9
End-to-end autonomous driving via Vision-Language-Action (VLA) models demands a precarious balance between high-fidelity trajectory planning and efficient inference. Existing parad…
SciCore-Mol: Augmenting Large Language Models with Pluggable Molecular Cognition Modules
Yuxuan Chen, Changwei Lv, Yunduo Xiao +5
Large Language Models (LLMs) are central to the one-for-all intelligent paradigm, but they face a fundamental challenge when dealing with heterogeneous scientific data such as mole…
StableVLA: Towards Robust Vision-Language-Action Models without Extra Data
Yiyang Fu, Chubin Zhang, Shukai Gong +7
It is infeasible to encompass all possible disturbances within the training dataset. This raises a critical question regarding the robustness of Vision-Language-Action (VLA) models…
HumanNet: Scaling Human-centric Video Learning to One Million Hours
Yufan Deng, Daquan Zhou
Progress in embodied intelligence increasingly depends on scalable data infrastructure. While vision and language have scaled with internet corpora, learning physical interaction r…
Enhancing Spatial Understanding in Image Generation via Reward Modeling
Zhenyu Tang, Chaoran Feng, Yufan Deng +5
Recent progress in text-to-image generation has greatly advanced visual fidelity and creativity, but it has also imposed higher demands on prompt complexity-particularly in encodin…
Rethinking Video Generation Model for the Embodied World
Yufan Deng, Zilin Pan, Hongyu Zhang +6
Video generation models have significantly advanced embodied intelligence, unlocking new possibilities for generating diverse robot data that capture perception, reasoning, and act…
MHLA: Restoring Expressivity of Linear Attention via Token-Level Multi-Head
Kewei Zhang, Ye Huang, Yufan Deng +5
While the Transformer architecture dominates many fields, its quadratic self-attention complexity hinders its use in large-scale applications. Linear attention offers an efficient…