COCO-2017

COCO 2017 image captions in Vietnamese The dataset is firstly introduced in dinhanhx/VisualRoBERTa. I use VinAI tools to translate COCO 2027 image caption (2017 Train/Val annotations) from English to Vietnamese. Then we merge UIT-ViIC dataset into it. To load the dataset, one can take a look at this code in VisualRoBER

Papers using COCO-2017 (4)

WaDi: Weight Direction-aware Distillation for One-step Image Synthesis2026Toward Lightweight and Fast Decoders for Diffusion Models in Image and Video Generation2025Grounding Text-to-Image Diffusion Models for Controlled High-Quality Image Generation2025Flash Diffusion: Accelerating Any Conditional Diffusion Model for Few Steps Image Generation2024