Research map / Architectures and efficiency / Neural architecture search
HyTAS: A Hyperspectral Image Transformer Architecture Search Benchmark and Analysis
Fangqin Zhou, Mert Kilickaya, Joaquin Vanschoren, Ran Piao
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 layer decay to ZiCo's score aggregation and use Fisher-style activation gradients (ZiCo++) | ZiCo · Indian Pines | Spearman correlation: 0.52 -> 0.73 | Improved |
| ZiCo++ instead of ZiCo | proxy search · PaviaU | test OA: 0.84 -> 0.84 | No clear effect |
| ZiCo++ instead of ZiCo | proxy search · Salinas | test OA: 0.91 -> 0.94 | Improved |
| use log(msa x mlp) instead of msa + mlp score (ablation) | Synflow · Indian Pines | Spearman correlation: 0.69 -> 0.72 | Improved |
| use log(msa x mlp) instead of msa + mlp score (ablation) | Synflow · PaviaU | Spearman correlation: 0.69 -> 0.63 | Hurt |
| use log(msa x mlp) instead of msa + mlp score (ablation) | SNIP · five HSI datasets | Spearman correlation: 0.70 -> 0.70 (Indian Pines) | No clear effect |
| use random input batch instead of real HSI data for proxies (ablation) | data-dependent proxies · Indian Pines | test OA / Spearman correlation: GradNorm 0.81 -> 0.81 | No clear effect |
| random input for JacobCov (ablation) | JacobCov · Indian Pines | Spearman correlation: 0.58 -> 0.68 (+0.1) | Improved |
| remove sign operations from Synflow, LogSynflow and DSS (ablation) | Synflow/LogSynflow/DSS · Indian Pines | search performance | Improved |
| train Random Forest on architecture features plus proxy scores to predict OA | proxy-based ranking · Indian Pines | Spearman correlation: 0.80 (9% above best proxy) | Improved |
| constrain search to larger model sizes (ablation) | proxies · Indian Pines | Spearman correlation | Hurt |
| choose proxies' top architecture instead of human-designed SpectralFormer | HSI transformer design · Indian Pines | test OA: 0.79 -> 0.81 | Improved |
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.
- ZiCo — ZiCo: Zero-shot NAS via inverse Coefficient of Variation on Gradients (ICLR 2023) · better on 4 of 5 shared results
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
- HyperSIGMA: Hyperspectral Intelligence Comprehension Foundation Model (TPAMI 2025 · similarity 0.86)
- VolFormer: Explore More Comprehensive Cube Interaction for Hyperspectral Image Restoration and Beyond (CVPR 2025 · similarity 0.85)
- Automatic Spectral Calibration of Hyperspectral Images: Method, Dataset and Benchmark (CVPR 2025 · similarity 0.84)
- Dual-Window Multiscale Transformer for Hyperspectral Snapshot Compressive Imaging (AAAI 2024 · similarity 0.84)
- Learning Multi-Modal Cross-Scale Deformable Transformer Network for Unregistered Hyperspectral Image Super-resolution (AAAI 2024 · similarity 0.84)
- SPECAT: SPatial-spEctral Cumulative-Attention Transformer for High-Resolution Hyperspectral Image Reconstruction (CVPR 2024 · similarity 0.84)
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.
