7 papers
Cosmos 3: Omnimodal World Models for Physical AI
NVIDIA, :, Aditi +293
We introduce Cosmos 3, a family of omnimodal world models designed to jointly process and generate language, image, video, audio, and action sequences within a unified mixture-of-t…
Are Reasoning Vision-Language Models Robust to Semantic Visual Distractions?
Yizheng Sun, Mochuan Zhan, Yanan Ma +10
Reasoning Vision-Language Models (VLMs) achieve strong performance on complex multimodal tasks, but reliable real-world application requires handling visual inputs that are messier…
LIMMT: Less is More for Motion Tracking
Yu Guan, Zekun Qi, Chenghuai Lin +7
We argue that high-quality motion data can steer tracking policies toward better optimization trajectories early in training. In this work, we introduce LIMMT (Less Is More for Mot…
Humanoid-GPT: Scaling Data and Structure for Zero-Shot Motion Tracking
Zekun Qi, Xuchuan Chen, Dairu Liu +10
We introduce Humanoid-GPT, a GPT-style Transformer with causal attention trained on a billion-scale motion corpus for whole-body control. Unlike prior shallow MLP trackers constrai…
A Self-Evolving Framework for Efficient Terminal Agents via Observational Context Compression
Jincheng Ren, Siwei Wu, Yizhi Li +8
As terminal agents scale to long-horizon, multi-turn workflows, a key bottleneck is not merely limited context length, but the accumulation of noisy terminal observations in the in…
Does Acceleration Cause Hidden Instability in Vision Language Models? Uncovering Instance-Level Divergence Through a Large-Scale Empirical Study
Yizheng Sun, Hao Li, Chang Xu +4
Vision-Language Models (VLMs) are powerful yet computationally intensive for widespread practical deployments. To address such challenge without costly re-training, post-training a…