content type videopublished August 12, 2025
Apple Workshop on Privacy-Preserving Machine Learning 2025: PREAMBLE: Private and Efficient Aggregation of Block Sparse Vectors and Applications
AuthorsHannah Keller (Aarhus University)
Related readings and updates.
One Layer Is Enough: Adapting Pretrained Visual Encoders for Image Generation
July 15, 2026research area Computer Visionconference CVPR
Visual generative models (e.g., diffusion models) typically operate in compressed latent spaces to balance training efficiency and sample quality. In parallel, there has been growing interest in leveraging high-quality pre-trained visual representations—either by aligning them inside VAEs or directly within the generative model. However, adapting such representations remains challenging due to fundamental mismatches between understanding-oriented…
CLaRa: Bridging Retrieval and Generation with Continuous Latent Reasoning
July 15, 2026research area Speech and Natural Language Processing
Retrieval-augmented generation (RAG) enhances large language models (LLMs) with external knowledge but still suffers from long contexts and disjoint retrieval–generation optimization. In this work, we propose CLaRa (Continuous Latent Reasoning), a unified framework that performs embedding-based compression and joint optimization in a shared continuous space. To obtain semantically rich and retrievable compressed vectors, thereby reducing the…
Discover opportunities in Machine Learning.
Our research in machine learning breaks new ground every day.