TY  - RPRT
TI  - GazeFormer-MoE: Context-Aware Gaze Estimation via CLIP and MoE Transformer
AU  - Xinyuan Zhao
AU  - Xianrui Chen
AU  - Ahmad Chaddad
PY  - 2026
UR  - https://arxiv.org/abs/2601.12316
ID  - 2601.12316
ER  -