9 stories tagged Transformers
A new arXiv study introduces Expert-Aware Contrast Decoding (EACD) for mixture-of-experts (MoE) models, showing it reduces AI hallucinations by leveraging expert-specific knowledge. The method outperforms standard contrastive decoding on transformer models, offering a path to more reliable AI assistants.
OpenAI has hired Noam Shazeer, a co-inventor of the Transformer technology, and Dean Ball, a former Trump AI policy official. These high-profile additions come as the company prepares for a possible public offering.
A new paper introduces the Integral Transform Network (ITNet), showing that convolutional networks, recurrent networks, and transformers are all variations of a single mathematical concept: a learnable integral transform. This unification could simplify AI development and lead to more versatile models.
PaddleOCR 3.5 is a free, open-source tool that can extract text and data from documents with high accuracy. It's now powered by Transformers, making it more powerful and easier to use for everyday tasks.
Researchers found that deep AI models can perform deductive reasoning nearly as well as models that follow step-by-step logic. This could make AI smarter without needing extra instructions.
Researchers have trained an AI model to generate pseudorandom numbers, mimicking how computers create seemingly random sequences. This could lead to more secure systems and new ways to test randomness.
Researchers introduce UniMatrix, a Universal Transformer variant that combines sparse retrieval with structured recurrence for efficient language modeling. The model achieves strong performance on associative recall tasks while maintaining computational efficiency.
Hugging Face has ported its Transformers library to MLX, enabling faster AI model inference on Apple Silicon. This move aims to leverage Apple's neural engine for enhanced performance.
OpenMythos is an open-source implementation of Claude Mythos, using a looped transformer with Mixture-of-Experts routing. This project aims to advance theoretical AI research by providing a transparent, modular framework.