CompART training improves VLM multi-object grounding and visual understanding

Researchers have developed a new training method called Compositional Attention-Regularized Training (CompART) to improve how Vision-Language Models (VLMs) handle complex, multi-object references. Current VLMs struggle with grounding performance when phrases involve multiple objects, largely due to training objectives that focus on image-caption alignment. CompART addresses this by decomposing captions into object-centric phrases and constructing composite phrases, encouraging the model's attention to balance across these components for better localization. AI

影响 Introduces a novel training technique to enhance VLM capabilities in understanding and localizing multiple objects within complex visual references.

排序理由 This is a research paper detailing a new training methodology for existing models. [lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.LG 阅读 →

• arXiv
• CompART
• Jiayun Luo
• Vision-Language Models

• 论文
• 模型发布

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →