Seonae Jeon

Seonae Jeon profile photo

Seonae Jeon

I am an undergraduate student in Computer Science and Engineering at DGIST.
My research interests broadly lie in Physical AI, with a particular focus on autonomous driving and the generalization of Vision-Language-Action models.

CV Email LinkedIn GitHub

Publications

Teaser for Unified Prediction and Planning via Conflict-Aware Disjoint Parameter Training
ECCV 2026 · Accepted

Unified Prediction and Planning via Conflict-Aware Disjoint Parameter Training

Taewon Seo1*, Seonae Jeon1*, Giwon Lee2*, Kuk-Jin Yoon2†, Daehee Park1†

A conflict-aware disjoint parameter training framework for unified prediction and planning, designed to reduce task interference and improve performance in multi-task autonomous driving models.

* Equal contribution   ·   † Corresponding authors

Project Page →