Saved in:
Bibliographic Details
Main Authors: Hou, Haowen, Huang, Zhiyi, Tan, Kaifeng, Lu, Rongchang, Yu, Fei Richard
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.21463
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913827887513600
author Hou, Haowen
Huang, Zhiyi
Tan, Kaifeng
Lu, Rongchang
Yu, Fei Richard
author_facet Hou, Haowen
Huang, Zhiyi
Tan, Kaifeng
Lu, Rongchang
Yu, Fei Richard
contents In this paper, we introduce RWKV-X, a novel hybrid architecture that combines the efficiency of RWKV for short-range modeling with a sparse attention mechanism designed to capture long-range context. Unlike previous hybrid approaches that rely on full attention layers and retain quadratic complexity, RWKV-X achieves linear-time complexity in training and constant-time complexity in inference decoding. We demonstrate that RWKV-X, when continually pretrained on 64K-token sequences, achieves near-perfect accuracy on the 64K passkey retrieval benchmark. It consistently outperforms prior RWKV-7 models on long-context benchmarks, while maintaining strong performance on short-context tasks. These results highlight RWKV-X as a scalable and efficient backbone for general-purpose language modeling, capable of decoding sequences up to 1 million tokens with stable speed and memory usage. To facilitate further research and analysis, we have made the checkpoints and the associated code publicly accessible at: https://github.com/howard-hou/RWKV-X.
format Preprint
id arxiv_https___arxiv_org_abs_2504_21463
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle RWKV-X: A Linear Complexity Hybrid Language Model
Hou, Haowen
Huang, Zhiyi
Tan, Kaifeng
Lu, Rongchang
Yu, Fei Richard
Computation and Language
In this paper, we introduce RWKV-X, a novel hybrid architecture that combines the efficiency of RWKV for short-range modeling with a sparse attention mechanism designed to capture long-range context. Unlike previous hybrid approaches that rely on full attention layers and retain quadratic complexity, RWKV-X achieves linear-time complexity in training and constant-time complexity in inference decoding. We demonstrate that RWKV-X, when continually pretrained on 64K-token sequences, achieves near-perfect accuracy on the 64K passkey retrieval benchmark. It consistently outperforms prior RWKV-7 models on long-context benchmarks, while maintaining strong performance on short-context tasks. These results highlight RWKV-X as a scalable and efficient backbone for general-purpose language modeling, capable of decoding sequences up to 1 million tokens with stable speed and memory usage. To facilitate further research and analysis, we have made the checkpoints and the associated code publicly accessible at: https://github.com/howard-hou/RWKV-X.
title RWKV-X: A Linear Complexity Hybrid Language Model
topic Computation and Language
url https://arxiv.org/abs/2504.21463