BlackGoose Rimer: Harnessing RWKV-7 as a Simple yet Superior Replacement for Transformers in Large-Scale Time Series Modeling

Li weile, Liu Xiao · 2025

Read BlackGoose Rimer: Harnessing RWKV-7 as a Simple yet Superior Replacement for Transformers in Large-Scale Time Series Modeling by Li weile, Liu Xiao (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-06121.txt.

Description

Li Weile and Liu Xiao present a 2025 paper on forecasting time series at large scale. Models for data that unfold over time, from energy use to traffic, struggle to grow the way language models have. Transformer designs are costly, and older recurrent nets have their own limits. The authors turn to RWKV-7, a newer design whose state update carries a meta-learning flavor.

They slot that design's two mixing parts into an existing transformer-based forecaster called Timer. The hybrid, nicknamed BlackGoose Rimer, improved results by a wide margin across their tests. It also trained about four and a half times faster while using a small fraction of the parameters. Code and model weights are posted openly for others to build on.

Catalog record

CREATORLi weile, Liu Xiao
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE14 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2503-06121

Browse the full library · About the Tower of Babel