A5102975841- Multimodal Machine Learning Applications
- Single-cell and spatial transcriptomics
- Advanced Image Processing Techniques
- Subtitles and Audiovisual Media
- Genomics and Chromatin Dynamics
- Language, Metaphor, and Cognition
- RNA modifications and cancer
- Image Processing Techniques and Applications
- Advanced Vision and Imaging
- Digital Storytelling and Education
- Digital Games and Media
University of Washington
2023
Shadow puppetry or shadow play, allows bodily participation into the process of linguistic storytelling, while potential multi-modal interaction through plays in existing large-language-model-based creative tools has not been fully discovered. We propose Narratron, a generative story-making tool that co-creates and co-performs children stories from using Claude 2 model. To achieve our system is designed to recognize hand gestural inputs as main character develop story plot accordance with...
Recent advancements in Multimodal Large Language Models (MM-LLMs) have demonstrated promising potential terms of generalization and robustness when applied to different modalities. While previous works already achieved 3D human motion generation using various approaches including language modeling, they mostly % are carefully designed use specialized architecture restricted single-human generation. Inspired by the success MM-LLMs, we propose MotionLLM, a simple general framework that can...
Abstract The rapid advancement of transposase-accessible chromatin using sequencing (ATAC-seq) technology, particularly with the emergence single-cell ATAC-seq (scATAC-seq), has accelerated studies gene regulation. However, absence a generic feature reference for data limits analyses and hinders development comprehensive cell atlases. To address this, we constructed accessibility by aggregating peaks from 624 high-quality bulk datasets, defining more than 1 million consensus (cPeaks)....