FigEx2: Visual-Conditioned Panel Detection and Captioning for Scientific Compound Figures
Published in EMNLP 2026 Main Conference (accepted), 2026
FigEx2 localizes panels and generates panel-wise captions directly from compound scientific figures. Access the paper.
Recommended citation: Jifeng Song, Arun Das, Pan Wang, Hui Ji, Kun Zhao, Yufei Huang, "FigEx2: Visual-Conditioned Panel Detection and Captioning for Scientific Compound Figures." EMNLP 2026 Main Conference, accepted.
Download Paper