Search Results for "linear"

...By introducing innovations such as matrix-based memory and improved normalization techniques, xLSTM improves the ability of recurrent networks to capture long-range dependencies in sequential data. The architecture aims to provide competitive performance with transformer-based models while maintaining advantages such as linear computational scaling and efficient memory usage for long sequences. Researchers have demonstrated that xLSTM models can scale to billions of parameters and large training datasets while maintaining efficient inference speeds.

MiniMax-01 is the official repository for two flagship models: MiniMax-Text-01, a long-context language model, and MiniMax-VL-01, a vision-language model built on top of it. MiniMax-Text-01 uses a hybrid attention architecture that blends Lightning Attention, standard softmax attention, and Mixture-of-Experts (MoE) routing to achieve both high throughput and long-context reasoning. It has 456 billion total parameters with 45.9 billion activated per token and is trained with advanced parallel...

Graph of Thoughts is an open-source framework that implements a novel reasoning paradigm for large language models by organizing reasoning steps as a structured graph instead of a simple linear chain. Traditional reasoning methods such as chain-of-thought generate sequential reasoning steps, but Graph of Thoughts introduces a more flexible structure where multiple reasoning paths can be explored and evaluated simultaneously. In this framework, problems are modeled as a graph of operations where nodes represent reasoning steps and edges represent dependencies between them. ...

Alpa is a system for training and serving large-scale neural networks. Scaling neural networks to hundreds of billions of parameters has enabled dramatic breakthroughs such as GPT-3, but training and serving these large-scale neural networks require complicated distributed system techniques. Alpa aims to automate large-scale distributed training and serving with just a few lines of code.

...In contrast, generalized additive models (GAMs) can maintain interpretability but often suffer from poor prediction performance due to their inability to effectively capture feature interactions. In this work, we aim to bridge this gap by using pre-trained neural language models to extract embeddings for each input before learning a linear model in the embedding space. The final model (which we call Emb-GAM) is a transparent, linear function of its input features and feature interactions. Leveraging the language model allows Emb-GAM to learn far fewer linear coefficients, model larger interactions, and generalize well to novel inputs. Across a variety of natural-language-processing datasets, Emb-GAM achieves strong prediction performance without sacrificing interpretability.