Modern deep learning methods typically treat image sequences as large tensors of sequentially stacked frames. However, is this straightforward representation id...
In hard-label black-box adversarial attacks, where only the top-1 predicted label is accessible, the prohibitive query complexity poses a major obstacle to prac...
State-of-the-art video generative models typically learn the distribution of video latents in the VAE space and map them to pixels using a VAE decoder. While th...
Recent advances in multimodal LLMs and systems that use tools for long-video QA point to the promise of reasoning over hour-long episodes. However, many methods...
Cognitive science suggests that spatial ability develops progressively-from perception to reasoning and interaction. Yet in multimodal LLMs (MLLMs), this hierar...
Current video avatar generation methods excel at identity preservation and motion alignment but lack genuine agency, they cannot autonomously pursue long-term g...
This paper proposes FedPOD (Proportionally Orchestrated Derivative) for optimizing learning efficiency and communication cost in federated learning among multip...
Point tracking aims to localize corresponding points across video frames, serving as a fundamental task for 4D reconstruction, robotics, and video editing. Exis...
We introduce Cube Bench, a Rubik's-cube benchmark for evaluating spatial and sequential reasoning in multimodal large language models (MLLMs). The benchmark dec...
Simulators can generate virtually unlimited driving data, yet imitation learning policies in simulation still struggle to achieve robust closed-loop performance...
Large vision-language models (VLMs) typically process hundreds or thousands of visual tokens per image or video frame, incurring quadratic attention cost and su...
Vision-language models (VLM) excel at general understanding yet remain weak at dynamic spatial reasoning (DSR), i.e., reasoning about the evolvement of object g...
