Research map / Image generation / Super-resolution
SRT: Super-Resolution for Time Series via Disentangled Rectified Flow
Jufang Duan, Shenglong Xiao, Yuren Zhang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add the implicit time function (ITF) (ablation) | SRT · ETTm1 SSR | MSE / DTW gap: +0.011 / +0.006 | Improved |
| add the implicit time function (ITF) (ablation) | SRT · weather SSR | MSE / DTW gap: +0.021 / +0.018 | Improved |
| add pattern smoothness in ITF (ablation) | SRT · PEMS-SF SSR | MSE / DTW gap: +0.017 / +0.015 | Improved |
| add the ITF multi-stage schema (ablation) | SRT · PEMS-SF SSR | MSE / DTW gap: +0.014 / +0.008 | Improved |
| add RoPE in the velocity predictor (ablation) | SRT · PEMS-SF SSR | MSE / DTW gap: +0.032 / +0.019 | Improved |
| add Pre-LN in the velocity predictor (ablation) | SRT · weather SSR | MSE / DTW gap: +0.018 / +0.011 | Improved |
| add the second CRA layer (attention to low-resolution series) (ablation) | SRT · weather SSR | MSE / DTW gap: +0.031 / +0.026 | Improved |
| add both cross-resolution attention layers (ablation) | SRT · weather SSR | MSE / DTW gap: +0.029 / +0.052 | Improved |
| disentangle detail and low-resolution into trend and periodic components (ablation) | SRT · PEMS-SF SSR | MSE / DTW gap: +0.046 / +0.073 | Improved |
| use the proposed velocity predictor instead of MLP (ablation) | SRT · PEMS-SF SSR | MSE gap: +0.203 | Improved |
| use the proposed velocity predictor instead of vanilla Transformer (ablation) | SRT · ETTm1 SSR | MSE / DTW gap: +0.085 / +0.048 | Improved |
| use TCN predictor instead of proposed predictor (ablation) | SRT · ETTm1 SSR | MSE / DTW gap: +0.002 / +0.001 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- ResShift — ResShift: Efficient Diffusion Model for Image Super-resolution by Residual Shifting (NeurIPS 2023) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Time Without Time: Pseudo-Temporal Representation for Space-Time Super-Resolution (CVPR 2026 · similarity 0.87)
- RSTT: Real-Time Spatial Temporal Transformer for Space-Time Video Super-Resolution (CVPR 2022 · similarity 0.86)
- NSARM: Next-Scale Autoregressive Modeling for Robust Real-World Image Super-Resolution (NeurIPS 2026 · similarity 0.85)
- SinSR: Diffusion-Based Image Super-Resolution in a Single Step (CVPR 2024 · similarity 0.85)
- VisionTS: Visual Masked Autoencoders Are Free-Lunch Zero-Shot Time Series Forecasters (ICML 2025 · similarity 0.85)
- EvEnhancer: Empowering Effectiveness, Efficiency and Generalizability for Continuous Space-Time Video Super-Resolution with Events (CVPR 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
