Recent Publications
Quickly discover relevant content by filtering publications.
From Pixels to Graphs: Open-Vocabulary Scene Graph Generation with Vision-Language Models
We introduce a new open-vocabulary SGG framework based on sequence generation. Our framework leverages vision-language pre-trained models (VLM) by incorporating an image-to-graph generation paradigm.
Learning by Correction: Efficient Tuning Task for Zero-Shot Generative Vision-Language Reasoning
We introduce Image-Conditioned Caption Correction (ICCC), a novel pre-training task designed to enhance VLMs’ zero-shot performance without the need for labeled task-aware data.
Mining Fine-Grained Image-Text Alignment for Zero-Shot Captioning via Text-Only Training
MacCap introduces a novel zero-shot image captioning framework with text-only training by noise injection training and visual subregion …
