Deep Learning
Robustly overfitting latents for flexible neural image compression
Neural image compression has made a great deal of progress. State-of-the-art models are based on variational autoencoders and are outperforming classical models.
Transformers are Minimax Optimal Nonparametric In-Context Learners Juno Kim 1,2 T ai Nakamaki 1 T aiji Suzuki 1,2 1 University of Tokyo 2
Large language models (LLMs) have demonstrated remarkable capabilities in understanding and generating natural language data.
BABILong: Testing the Limits of LLMs with Long Context Reasoning-in-a-Haystack
To bridge this gap, we introduce the BABILong benchmark, designed to test language models' ability to reason across facts distributed in extremely
No Filter: Cultural and Socioeconomic Diversity in Contrastive Vision-Language Models Angéline Pouget
Second, pretraining with global, unfiltered data before fine-tuning on English content can improve cultural understanding without sacrificing performance on said popular benchmarks.
