PIE Text Encoder: Padding Is Enough to generate Image in diffusion based Text-to-Image Model | IEEE Conference Publication | IEEE Xplore