{"ID":2858425,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.08771","arxiv_id":"2510.08771","title":"LinearSR: Unlocking Linear Attention for Stable and Efficient Image Super-Resolution","abstract":"Generative models for Image Super-Resolution (SR) are increasingly powerful, yet their reliance on self-attention's quadratic complexity (O(N^2)) creates a major computational bottleneck. Linear Attention offers an O(N) solution, but its promise for photorealistic SR has remained largely untapped, historically hindered by a cascade of interrelated and previously unsolved challenges. This paper introduces LinearSR, a holistic framework that, for the first time, systematically overcomes these critical hurdles. Specifically, we resolve a fundamental, training instability that causes catastrophic model divergence using our novel \"knee point\"-based Early-Stopping Guided Fine-tuning (ESGF) strategy. Furthermore, we mitigate the classic perception-distortion trade-off with a dedicated SNR-based Mixture of Experts (MoE) architecture. Finally, we establish an effective and lightweight guidance paradigm, TAG, derived from our \"precision-over-volume\" principle. Our resulting LinearSR model simultaneously delivers state-of-the-art perceptual quality with exceptional efficiency. Its core diffusion forward pass (1-NFE) achieves SOTA-level speed, while its overall multi-step inference time remains highly competitive. This work provides the first robust methodology for applying Linear Attention in the photorealistic SR domain, establishing a foundational paradigm for future research in efficient generative super-resolution.","short_abstract":"Generative models for Image Super-Resolution (SR) are increasingly powerful, yet their reliance on self-attention's quadratic complexity (O(N^2)) creates a major computational bottleneck. Linear Attention offers an O(N) solution, but its promise for photorealistic SR has remained largely untapped, historically hindered...","url_abs":"https://arxiv.org/abs/2510.08771","url_pdf":"https://arxiv.org/pdf/2510.08771v4","authors":"[\"Xiaohui Li\",\"Shaobin Zhuang\",\"Shuo Cao\",\"Yang Yang\",\"Yuandong Pu\",\"Qi Qin\",\"Siqi Luo\",\"Bin Fu\",\"Yihao Liu\"]","published":"2025-10-09T19:41:51Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Mixture of Experts\",\"Diffusion Model\"]","has_code":false}