← Search

Byungwoo Jeon

2 accepted papers

2026

Vision-aligned Latent Reasoning for Multi-Modal Large Language Model

ICML 2026poster

Despite recent advancements in Multi-modal Large Language Models (MLLMs) on diverse understanding tasks, these models struggle to solve problems which require extensive multi-step reasoning. This is primarily due to the progressive dilution of visual information during long-context generation, which…

Cited by 0SourceScholar
2024

TrackIME: Enhanced Video Point Tracking via Instance Motion Estimation

NeurIPS 2024spotlight

Tracking points in video frames is essential for understanding video content. However, the task is fundamentally hindered by the computation demands for brute-force correspondence matching across the frames. As the current models down-sample the frame resolutions to mitigate this challenge, they fal…

Cited by 0SourcePDFScholar