Abstract: This paper introduces SparseTSF, a novel and extremely lightweight method for Long-term Time Series Forecasting (LTSF), designed to address the challenges of modeling complex temporal ...

Abstract: Self-supervised learning (SSL) is a data-driven learning approach that utilizes the innate structure of the data to guide the learning process. In contrast to supervised learning, which ...

A lightweight framework that gives language models (LMs) a persistent, evolving memory during inference time. Dynamic Cheatsheet (DC) endows black-box language models with the ability to store and ...

Discover how Fourier Analysis breaks down complex time series data into simpler components to identify trends and patterns, despite its limitations in stock forecasting.

Xoxe was introduced as an advanced AI robot designed to learn faster than previous systems. What caught attention wasn’t just its intelligence, but its ability to predict events unfolding in the real ...

We introduce PaCoRe (Parallel Coordinated Reasoning), a framework that shifts the driver of inference from sequential depth to coordinated parallel breadth, breaking the model context limitation and ...