
Selective attention-driven modulation can effectively enhance the performance and robustness of continual learning models by leveraging the forgetting-free behavior of saliency prediction.


coremsg

leveraging-selective-attention-for-robust-continual-learning


Leveraging Selective Attention for Robust Continual Learning


title_rewrite


Language models can maintain a majority of the performance of vision-language models in learning new visual tasks from limited data, suggesting language plays a key role in visual understanding.


exploring-the-contributions-of-language-and-vision-to-learning-from-limited-data


Exploring the Contributions of Language and Vision to Learning from Limited Data



Neural networks can rapidly and accurately predict the communication and navigation attributes of swarming autonomous agents, enabling real-time inference of their overall tactics.


rapid-inference-of-swarm-tactics-and-attributes-using-neural-networks


Rapid Inference of Swarm Tactics and Attributes Using Neural Networks
