4 papers
Your Embedding Model is SMARTer Than You Think
Jianrui Zhang, Hyun Jung Lee, Sukanta Ganguly +3
Multimodal retrieval relies heavily on single-vector retrievers, which compress rich, sequential token sequences into one single global representation. While efficient, they discar…
Latent Recurrent Transformer: Architecture Exploration, Training Strategies, and Scaling Behavior
Zeyi Huang, Xuehai He, LiLiang Ren +8
We study Latent Recurrent Transformer (LRT), a lightweight augmentation of autoregressive transformers that reuses a high-level source-layer hidden state from the previous token as…
Talk in Pieces, See in Whole: Disentangled and Hierarchical Representation Learning in Language-based Object Detection
Sojung An, Kwanyong Park, Yong Jae Lee +1
Vision-language models (VLMs) have advanced multimodal perception, demonstrated by open-vocabulary object detection with simple language queries. State-of-the-art VLMs still strugg…
MATE: Meet At The Embedding -- Connecting Images with Long Texts
Young Kyun Jang, Junmo Kang, Yong Jae Lee +1
While advancements in Vision Language Models (VLMs) have significantly improved the alignment of visual and textual data, these models primarily focus on aligning images with short…