Research map / Architectures and efficiency / RNNs and normalization
$(\textrm{Implicit})^2$: Implicit Layers for Implicit Representations
Zhichun Huang, Shaojie Bai, J Zico Kolter
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 |
|---|---|---|---|
| replace explicit 1L-512D with implicit 1L-512D layer | Fourier-MFN · Image generalization, Natural | PSNR (dB): 24.57 -> 25.19 | Improved |
| replace explicit 4L-256D with implicit 4L-256D... (1L implicit vs 4L explicit comparison) | Fourier-MFN · Image generalization, Text | PSNR (dB): 24.84 -> 27.19 | Mixed |
| make Gabor-MFN implicit | Gabor-MFN · Image generalization, Text, 1L-512D | PSNR (dB): 27.74 -> 28.07 | Improved |
| make SIREN implicit (iSIREN) | SIREN (input inj.) · Image generalization, Natural, 1L-256D | PSNR (dB): 22.88 -> 24.28 | Improved |
| make implicit models 4 layers deep (4L-256D) on image generalization | Fourier-iMFN · Image generalization, Natural | PSNR (dB): 24.52 -> 24.57 | No clear effect |
| make Fourier-MFN implicit (1L-512D) for audio | Fourier-MFN · Audio representation | PSNR (dB): 35.42 -> 43.79 | Improved |
| use deeper implicit layer (4L-256D) for audio | Fourier-iMFN · Audio representation | PSNR (dB): 43.79 -> 47.29 | Improved |
| make Fourier-MFN implicit for video, 1L-1024D | Fourier-MFN · Video representation | PSNR (dB): 24.97 -> 25.85 | Improved |
| make Gabor-MFN implicit for video, 4L-1024D | Gabor-MFN · Video representation | PSNR (dB): 29.64 -> 29.20 | Hurt |
| make Fourier-MFN implicit for 3D occupancy, 1L-512D | Fourier-MFN · 3D occupancy, dragon | IoU: 0.9650 -> 0.9698 | Improved |
| use 0th-order gradient approximation (T = 0) instead of T = 1 (ablation) | Fourier-iMFN 1L-512D · Image representation 512x512 | training PSNR | Hurt |
| use T = 1 truncated gradient instead of T = 0 (ablation) | Fourier-iMFN 1L-512D · Image representation 512x512 | training PSNR | Improved |
Closest papers by meaning
- VI^3NR: Variance Informed Initialization for Implicit Neural Representations (CVPR 2025 · similarity 0.86)
- FreSh: Frequency Shifting for Accelerated Neural Representation Learning (ICLR 2025 · similarity 0.86)
- MIRE: Matched Implicit Neural Representations (CVPR 2025 · similarity 0.86)
- Implicit Zoo: A Large-Scale Dataset of Neural Implicit Functions for 2D Images and 3D Scenes (NeurIPS 2024 · similarity 0.85)
- Implicit regularization of deep residual networks towards neural ODEs (ICLR 2024 · similarity 0.85)
- Few-shot Implicit Function Generation via Equivariance (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.
