Pith. sign in

REVIEW 20 cited by

Adaptive Fourier Neural Operators: Efficient Token Mixers for Transformers

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2111.13587 v2 pith:TYPLYCY3 submitted 2021-11-24 cs.CV cs.LG

Adaptive Fourier Neural Operators: Efficient Token Mixers for Transformers

classification cs.CV cs.LG
keywords afnofourierlearningtokenefficientmixingadaptiveconvolution
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Vision transformers have delivered tremendous success in representation learning. This is primarily due to effective token mixing through self attention. However, this scales quadratically with the number of pixels, which becomes infeasible for high-resolution inputs. To cope with this challenge, we propose Adaptive Fourier Neural Operator (AFNO) as an efficient token mixer that learns to mix in the Fourier domain. AFNO is based on a principled foundation of operator learning which allows us to frame token mixing as a continuous global convolution without any dependence on the input resolution. This principle was previously used to design FNO, which solves global convolution efficiently in the Fourier domain and has shown promise in learning challenging PDEs. To handle challenges in visual representation learning such as discontinuities in images and high resolution inputs, we propose principled architectural modifications to FNO which results in memory and computational efficiency. This includes imposing a block-diagonal structure on the channel mixing weights, adaptively sharing weights across tokens, and sparsifying the frequency modes via soft-thresholding and shrinkage. The resulting model is highly parallel with a quasi-linear complexity and has linear memory in the sequence size. AFNO outperforms self-attention mechanisms for few-shot segmentation in terms of both efficiency and accuracy. For Cityscapes segmentation with the Segformer-B3 backbone, AFNO can handle a sequence size of 65k and outperforms other efficient self-attention mechanisms.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 20 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Structure Before Collapse: Transient semantic geometry in next-token prediction

    cs.LG 2026-06 unverdicted novelty 7.0

    Semantic geometry emerges transiently early in next-token prediction training before collapsing to Neural Collapse symmetry in synthetic settings with latent semantic factors.

  2. CHASM: Cross-frequency Harmonized Axis-Separable Mixing for Spectral Token Operators

    cs.CV 2026-05 unverdicted novelty 7.0

    CHASM introduces a cross-frequency harmonized axis-separable spectral mixer using a shared channel eigenbasis plus per-frequency positive gains, yielding consistent gains over same-backbone baselines in medical and na...

  3. Latent Fourier Transform

    cs.SD 2026-04 unverdicted novelty 7.0

    LatentFT uses latent-space Fourier transforms and frequency masking in diffusion autoencoders to enable timescale-specific manipulation of musical structure in generative models.

  4. Non-markovian neural quantum propagator and its application to the simulation of ultrafast nonlinear spectra

    physics.chem-ph 2024-08 unverdicted novelty 7.0

    A machine learning model called neural quantum propagator is introduced to efficiently solve non-Markovian quantum dynamics described by HEOM and applied to simulate spectra of the FMO complex.

  5. Chiaroscuro Attention: Spending Compute in the Dark

    cs.CL 2026-06 unverdicted novelty 6.0

    CHIAR-Former routes tokens via spectral entropy to DCT mixing or attention, yielding 35-40% FLOP savings at 400M parameters with modest perplexity increase on WikiText-103.

  6. Uncovering Extreme Event Mechanisms for Prediction and Control with Sensitivity-Balanced Projections

    nlin.CD 2026-06 unverdicted novelty 6.0

    CoBRAS projections via backpropagation capture sensitivity to enable data-driven prediction and control of extreme events in chaotic dynamical systems including turbulence, oscillator networks, and rogue waves.

  7. Skillful high-resolution weather forecasting independent of physical models

    physics.ao-ph 2026-05 unverdicted novelty 6.0

    ObsCast produces skillful short-term high-resolution weather analyses and forecasts over the contiguous US and Europe using only observational data, outperforming operational NWP without relying on NWP-derived data fo...

  8. AOT-POT: Adaptive Operator Transformation for Large-Scale PDE Pre-training

    cs.LG 2026-05 unverdicted novelty 6.0

    AOT-POT adaptively reshapes complex PDE solution operators via input-dependent transformations and parallel stream mixing to enable effective large-scale pre-training, yielding SOTA results on 12 benchmarks with minim...

  9. U-HNO: A U-shaped Hybrid Neural Operator with Sparse-Point Adaptive Routing for Non-stationary PDE Dynamics

    cs.LG 2026-05 unverdicted novelty 6.0

    U-HNO uses adaptive per-point routing in a U-shaped hybrid architecture to achieve state-of-the-art accuracy on PDE benchmarks with sharp localized features.

  10. Don't Fix the Basis -- Learn It: Spectral Representation with Adaptive Basis Learning for PDEs

    cs.LG 2026-05 unverdicted novelty 6.0

    ABLE learns a spatially adaptive Parseval frame from data via an ancillary density to replace fixed bases in spectral neural operators for PDEs.

  11. Large-eddy simulation nets (LESnets) based on physics-informed neural operator for wall-bounded turbulence

    physics.flu-dyn 2026-04 unverdicted novelty 6.0

    LESnets integrates LES equations and the law of the wall into F-FNO to enable data-free, stable long-term predictions of wall-bounded turbulence at Re_tau up to 1000 on coarse grids, matching traditional LES accuracy ...

  12. Earth System Foundation Model (ESFM): A unified framework for heterogeneous data integration and forecasting

    physics.ao-ph 2026-04 unverdicted novelty 6.0

    ESFM is a single open foundation model that unifies heterogeneous Earth data sources and forecasts missing regions while preserving inter-variable physical relationships.

  13. Multi-Scale Wavelet Transformers for Operator Learning of Dynamical Systems

    cs.LG 2026-02 unverdicted novelty 6.0

    Multi-scale wavelet transformers learn operator dynamics of chaotic systems in the wavelet domain, yielding lower errors and higher spectral fidelity on benchmarks and ERA5 climate data.

  14. NOWS: Neural Operator Warm Starts for Accelerating Iterative Solvers

    cs.LG 2025-11 unverdicted novelty 6.0

    Neural operators supply warm-start guesses that cut iteration counts and runtime by up to 90% in Krylov solvers for PDEs while retaining the original methods' convergence guarantees.

  15. Estimating Parameter Fields in Multi-Physics PDEs from Scarce Measurements

    cs.LG 2025-08 unverdicted novelty 6.0

    Neptune infers spatiotemporal parameter fields in PDEs from as few as 45 sparse measurements using independent coordinate neural networks, outperforming PINNs and neural operators with lower errors and better extrapolation.

  16. Foundations of Practical Quantum Advantage in Quantum-Informed Machine Learning for Predicting Chaos

    quant-ph 2026-06 unverdicted novelty 5.0

    Develops k-indexed Q-Priors for quantum-informed ML on chaotic systems, proving a two-stage quantum-classical separation in measurement complexity and demonstrating it in turbulent flow and weather forecasting workflows.

  17. Caracal: Causal Architecture via Spectral Mixing

    cs.LG 2026-04 unverdicted novelty 5.0

    Caracal is a Fourier-based sequence mixing architecture that achieves causal autoregressive modeling with standard operators and competitive performance on long sequences.

  18. Euler-inspired Decoupling Neural Operator for Efficient Pansharpening

    cs.CV 2026-04 unverdicted novelty 5.0

    EDNO redefines pansharpening as a frequency-domain functional mapping that decouples fusion via Euler-inspired polar coordinates into explicit phase-rotation simulation and implicit spectral modeling for improved efficiency.

  19. Physics-guided spatiotemporal neural models for fuel density prediction

    cs.LG 2026-07 conditional novelty 4.0

    Adding physics-guided loss terms to ConvLSTM, AFNONet, and ViViT improves fuel density prediction accuracy and stability over purely data-driven baselines on simulated prescribed-fire data.

  20. A Geometry-Aware Triplane Field Network for Vehicle Aerodynamic Prediction

    cs.LG 2026-06 unverdicted novelty 4.0

    GTF-Net combines triplane features, AFNO spectral mixing, CNN refinement, and explicit geometric encodings to reduce relative L2 error on vehicle pressure and shear stress prediction versus prior baselines.