no code implementations • 29 Aug 2023 • Yung-Hsin Chen, Yuli Zhou
The study investigates the potential of post-OCR models to overcome limitations in OCR models and explores the impact of incorporating glyph embedding on post-OCR correction performance.
no code implementations • 28 Aug 2023 • Yuansheng Ni, Sichao Jiang, Xinyu Wu, Hui Shen, Yuli Zhou
The focus is on the robustness of instruction-tuned LLMs to seen and unseen tasks.