Deep Learning
Towards Open-Vocabulary Semantic Segmentation Without Semantic Labels Heeseong Shin 1 Chaehyun Kim
Large-scale vision-language models like CLIP have demonstrated impressive open-vocabulary capabilities for image-level tasks, excelling in recognizing what objects are present.
AlphaPruning: Using Heavy-Tailed Self Regularization Theory for Improved Layer-wise Pruning of Large Language Models
We have open-sourced our code.
Unlearning in Open-Source LLMs through the Embedding Space
LLMs that exploit full model access remain largely unexplored.
Improving Deep Learning Optimization through Constrained Parameter Regularization
Regularization is a critical component in deep learning.
10e6dfea9a673bef4a7b1cb9234891bc-Paper-Conference.pdf
The pretraining data of today's strongest language models is opaque; in particular,
MMDU: A Multi-T urn Multi-Image Dialog Understanding Benchmark and Instruction-T uning Dataset for L VLMs
Generating natural and meaningful responses to communicate with multi-modal human inputs is a fundamental capability of Large Vision-Language Models (L VLMs).