# MRT: Learning Compact Representations with Mixed RWKV-Transformer for Extreme Image Compression

arXiv: https://arxiv.org/abs/2511.06717 · HTML: https://mlsanity.com/paper/2511.06717/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 37 references that are in the map.
- First posted on arXiv 10 Nov 2025.

## Builds on (most cited first, 24 of 24)

- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Mamba: Linear-Time Sequence Modeling with Selective State Spaces](https://mlsanity.com/paper/2312.00752/) (2023)
- [Taming Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2012.09841/) (2020)
- [Variational image compression with a scale hyperprior](https://mlsanity.com/paper/1802.01436/) (2018)
- [RWKV: Reinventing RNNs for the Transformer Era](https://mlsanity.com/paper/2305.13048/) (2023)
- [Image Quality Assessment: Unifying Structure and Texture Similarity](https://mlsanity.com/paper/2004.07728/) (2020)
- [Language Model Beats Diffusion -- Tokenizer is Key to Visual Generation](https://mlsanity.com/paper/2310.05737/) (2023)
- [Learned Image Compression with Discretized Gaussian Mixture Likelihoods and Attention Modules](https://mlsanity.com/paper/2001.01568/) (2020)
- [High-Fidelity Generative Image Compression](https://mlsanity.com/paper/2006.09965/) (2020)
- [An Image is Worth 32 Tokens for Reconstruction and Generation](https://mlsanity.com/paper/2406.07550/) (2024)
- [Rethinking Lossy Compression: The Rate-Distortion-Perception Tradeoff](https://mlsanity.com/paper/1901.07821/) (2019)
- [Learned Image Compression with Mixed Transformer-CNN Architectures](https://mlsanity.com/paper/2303.14978/) (2023)
- [Lossy Image Compression with Conditional Diffusion Models](https://mlsanity.com/paper/2209.06950/) (2022)
- [FlexTok: Resampling Images into 1D Token Sequences of Flexible Length](https://mlsanity.com/paper/2502.13967/) (2025)
- [Vision-RWKV: Efficient and Scalable Visual Perception with RWKV-Like Architectures](https://mlsanity.com/paper/2403.02308/) (2024)
- [Improving Statistical Fidelity for Neural Image Compression with Implicit Local Likelihood Models](https://mlsanity.com/paper/2301.11189/) (2023)
- [Towards Extreme Image Compression with Latent Feature Guidance and Diffusion Prior](https://mlsanity.com/paper/2404.18820/) (2024)
- [Restore-RWKV: Efficient and Effective Medical Image Restoration with RWKV](https://mlsanity.com/paper/2407.11087/) (2024)
- [Flow to the Mode: Mode-Seeking Diffusion Autoencoders for State-of-the-Art Image Tokenization](https://mlsanity.com/paper/2503.11056/) (2025)
- [Extreme Image Compression using Fine-tuned VQGANs](https://mlsanity.com/paper/2307.08265/) (2023)
- [RDEIC: Accelerating Diffusion-Based Extreme Image Compression with Relay Residual Diffusion](https://mlsanity.com/paper/2410.02640/) (2024)
- [Unifying Generation and Compression: Ultra-low bitrate Image Coding Via Multi-stage Transformer](https://mlsanity.com/paper/2403.03736/) (2024)
- [Tree-Structured Data Clustering-Driven Neural Network for Intra Prediction in Video Coding](https://mlsanity.com/paper/2106.05481/) (2021)

## Built on by (most cited first, 1 of 1)

- [MTLSI-Net: A Linear Semantic Interaction Network for Parameter-Efficient Multi-Task Dense Prediction](https://mlsanity.com/paper/2604.01995/) (2026)

## Abstract (arXiv)

Recent advances in extreme image compression have revealed that mapping pixel data into highly compact latent representations can significantly improve coding efficiency. However, most existing methods compress images into 2-D latent spaces via convolutional neural networks (CNNs) or Swin Transformers, which tend to retain substantial spatial redundancy, thereby limiting overall compression performance. In this paper, we propose a novel Mixed RWKV-Transformer (MRT) architecture that encodes images into more compact 1-D latent representations by synergistically integrating the complementary strengths of linear-attention-based RWKV and self-attention-based Transformer models. Specifically, MRT partitions each image into fixed-size windows, utilizing RWKV modules to capture global dependencies across windows and Transformer blocks to model local redundancies within each window. The hierarchical attention mechanism enables more efficient and compact representation learning in the 1-D domain. To further enhance compression efficiency, we introduce a dedicated RWKV Compression Model (RCM) tailored to the structure characteristics of the intermediate 1-D latent features in MRT. Extensive experiments on standard image compression benchmarks validate the effectiveness of our approach. The proposed MRT framework consistently achieves superior reconstruction quality at bitrates below 0.02 bits per pixel (bpp). Quantitative results based on the DISTS metric show that MRT significantly outperforms the state-of-the-art 2-D architecture GLC, achieving bitrate savings of 43.75%, 30.59% on the Kodak and CLIC2020 test datasets, respectively.