NTK-Guided Implicit Neural Teaching
Not on the map. Why?
Abstract (arXiv)
Implicit Neural Representations (INRs) parameterize continuous signals via multilayer perceptrons (MLPs), enabling compact, resolution-independent modeling for tasks like image, audio, and 3D reconstruction. However, fitting high-resolution signals demands optimizing over millions of coordinates, incurring prohibitive computational costs. To address it, we propose NTK-Guided Implicit Neural Teaching (NINT), which accelerates training by dynamically selecting coordinates that maximize global functional updates. Leveraging the Neural Tangent Kernel (NTK), NINT scores examples by the norm of their NTK-augmented loss gradients, capturing both fitting errors and heterogeneous leverage (self-influence and cross-coordinate coupling). This dual consideration enables faster convergence compared to existing methods. Through extensive experiments, we demonstrate that NINT significantly reduces training time by nearly half while maintaining or improving representation quality, establishing state-of-the-art acceleration among recent sampling-based strategies.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 59 of its 75 references that are in the map. First posted on arXiv 19 Nov 2025.
Builds on
59 of 75 references.
References · 59
- 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
- 2022Instant Neural Graphics Primitives with a Multiresolution Hash Encodingcited by 3,698
- 2019DeepSDF: Learning Continuous Signed Distance Functions for Shape Representationcited by 2,682
- 2018Neural Tangent Kernel: Convergence and Generalization in Neural Networkscited by 2,597
- 2020Implicit Neural Representations with Periodic Activation Functionscited by 2,214
- 2020Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domainscited by 2,170
- 2018On the Spectral Bias of Neural Networkscited by 1,321
- 2022TensoRF: Tensorial Radiance Fieldscited by 1,275
- 2020NeRF in the Wild: Neural Radiance Fields for Unconstrained Photo Collectionscited by 1,230
- 2016An overview of gradient descent optimization algorithmscited by 1,021
- 2020Neural Sparse Voxel Fieldscited by 959
- 2019Wide Neural Networks of Any Depth Evolve as Linear Models Under Gradient Descentcited by 732
38 more
- 2020Implicit Geometric Regularization for Learning Shapescited by 707
- 2021KiloNeRF: Speeding up Neural Radiance Fields with Thousands of Tiny MLPscited by 672
- 2019On Exact Computation with an Infinitely Wide Neural Netcited by 572
- 2022Point-NeRF: Point-based Neural Radiance Fieldscited by 503
- 2023Neuralangelo: High-Fidelity Neural Surface Reconstructioncited by 464
- 2018SDR - half-baked or well done?cited by 420
- 2019SAL: Sign Agnostic Learning of Shapes from Raw Datacited by 389
- 2020Deep Local Shapes: Learning Local SDF Priors for Detailed 3D Reconstructioncited by 321
- 2021NeRV: Neural Representations for Videoscited by 255
- 2020Learned Initializations for Optimizing Coordinate-Based Neural Representationscited by 240
- 2021COIN: COmpression with Implicit Neural representationscited by 232
- 2023WIRE: Wavelet Implicit Neural Representationscited by 220
- 2021Implicit Neural Representations for Image Compressioncited by 161
- 2021BACON: Band-limited Coordinate Networks for Multiscale Scene Representationcited by 142
- 2021Modulated Periodic Activations for Generalizable Local Functional Representationscited by 140
- 2021Beyond Periodicity: Towards a Unifying Framework for Activations in Coordinate-MLPscited by 134
- 2021A Structured Dictionary Perspective on Implicit Neural Representationscited by 100
- 2023Implicit Neural Representation in Medical Imaging: A Comparative Surveycited by 88
- 2022Transformers as Meta-Learners for Implicit Neural Representationscited by 79
- 2022NIRVANA: Neural Implicit Representations of Videos with Adaptive Networks and Autoregressive Patch-wise Modelingcited by 50
- 2023CopyRNeRF: Protecting the CopyRight of Neural Radiance Fieldscited by 33
- 2022Generalised Implicit Neural Representationscited by 31
- 2022Compressing multidimensional weather and climate data into neural networkscited by 31
- 2022DINER: Disorder-Invariant Implicit Neural Representationcited by 30
- 2022Scalable Neural Video Representations with Learnable Positional Featurescited by 29
- 2023Modality-Agnostic Variational Compression of Implicit Neural Representationscited by 22
- 2021A Multi-Implicit Neural Representation for Fontscited by 21
- 2024Protecting NeRFs' Copyright via Plug-And-Play Watermarking Base Modelcited by 18
- 2024DDMI: Domain-Agnostic Latent Diffusion Models for Synthesizing High-Quality Implicit Neural Representationscited by 16
- 2023Regularize implicit neural representation by itselfcited by 13
- 2023Partition Speeds Up Learning Implicit Neural Representations Based on Exponential-Increase Hypothesiscited by 13
- 2023Accelerating Neural Field Training via Soft Miningcited by 13
- 2024SketchINR: A First Look into Sketches as Implicit Neural Representationscited by 13
- 2024Nonparametric Teaching of Implicit Neural Representationscited by 13
- 2023Hypernetworks build Implicit Neural Representations of Soundscited by 11
- 2023Learning Spatially Collaged Fourier Bases for Implicit Neural Representationcited by 11
- 2023Inverting the Imaging Process by Learning an Implicit Camera Modelcited by 10
- 2025The NeRF Signature: Codebook-Aided Watermarking for Neural Radiance Fieldscited by 10
50 most cited of 59 All 59 (JSON)
Built on by
Citing papers per year: 2026: 2.
2 papers.