Pith. sign in

REVIEW 18 cited by

U-KAN Makes Strong Backbone for Medical Image Segmentation and Generation

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

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 2406.02918 v3 pith:ADWJY34T submitted 2024-06-05 eess.IV cs.CV

Chenxin Li , Xinyu Liu , Wuyang Li , Cheng Wang , Hengyu Liu , Yifan Liu , Zhen Chen , Yixuan Yuan This is my paper · ORCID
classification eess.IVcs.CV
keywords u-kanimagesegmentationmedicalu-netaccuracybackbonediffusion
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

U-Net has become a cornerstone in various visual applications such as image segmentation and diffusion probability models. While numerous innovative designs and improvements have been introduced by incorporating transformers or MLPs, the networks are still limited to linearly modeling patterns as well as the deficient interpretability. To address these challenges, our intuition is inspired by the impressive results of the Kolmogorov-Arnold Networks (KANs) in terms of accuracy and interpretability, which reshape the neural network learning via the stack of non-linear learnable activation functions derived from the Kolmogorov-Anold representation theorem. Specifically, in this paper, we explore the untapped potential of KANs in improving backbones for vision tasks. We investigate, modify and re-design the established U-Net pipeline by integrating the dedicated KAN layers on the tokenized intermediate representation, termed U-KAN. Rigorous medical image segmentation benchmarks verify the superiority of U-KAN by higher accuracy even with less computation cost. We further delved into the potential of U-KAN as an alternative U-Net noise predictor in diffusion models, demonstrating its applicability in generating task-oriented model architectures. These endeavours unveil valuable insights and sheds light on the prospect that with U-KAN, you can make strong backbone for medical image segmentation and generation. Project page:\url{https://yes-u-kan.github.io/}.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 18 Pith papers

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

  1. KANEL\'E: Kolmogorov-Arnold Networks for Efficient LUT-based Evaluation

    cs.AR 2025-12 conditional novelty 7.0 of 10

    Quantized, pruned Kolmogorov-Arnold Networks can be compiled directly into FPGA lookup tables, achieving extreme latency/resource reductions and matching state-of-the-art LUT-based networks on several benchmarks.

  2. KAN-SAs: Efficient Acceleration of Kolmogorov-Arnold Networks on Systolic Arrays

    cs.AR 2025-11 conditional novelty 7.0 of 10

    A systolic-array accelerator that tabulates B-splines and exploits B-spline local support achieves ~100% PE utilization and a 2x cycle reduction for KAN inference compared with a conventional systolic array.

  3. MetaScope: Optics-Driven Neural Network for Ultra-Micro Metalens Endoscopy

    cs.CV 2025-08 unverdicted novelty 6.0 of 10

    MetaScope, an optics-driven network, corrects metalens endoscope images and outperforms prior methods on segmentation and restoration.

  4. Improving Memory Efficiency for Training KANs via Meta Learning

    cs.LG 2025-06 conditional novelty 6.0 of 10

    MetaKANs generates each KAN activation function from a shared prompt-conditioned meta-learner, cutting trainable parameters toward MLP level while retaining comparable or better accuracy on tested benchmarks.

  5. PRKAN: Parameter-Reduced Kolmogorov-Arnold Networks

    cs.LG 2025-01 conditional novelty 6.0 of 10

    PRKAN lowers KAN parameter counts to near-MLP levels via attention, convolution/pooling, dimension summation, and feature-vector projections, reaching MLP-like accuracy on MNIST and Fashion-MNIST.

  6. LWFNet: Coherent Doppler Wind Lidar-Based Network for Wind Field Retrieval

    physics.ao-ph 2025-01 reject novelty 6.0 of 10

    LWFNet, a Transformer plus KAN network, retrieves wind fields directly from coherent Doppler lidar spectra and reportedly beats the spectral centroid baseline, but the reported gains rely on a test-set-fitted bias correction.

  7. Quantum Variational Activation Functions Empower Kolmogorov-Arnold Networks

    quant-ph 2025-09 reject novelty 5.0 of 10

    QKANs show strong empirical performance on regression, vision, and language tasks, but the claimed exponential parameter reduction is not rigorously established.

  8. Track Any Anomalous Object: A Granular Video Anomaly Detection Pipeline

    cs.CV 2025-06 conditional novelty 5.0 of 10

    TAO pipelines object-centric anomaly scores into SAM2 prompts with a temporal consistency filter to obtain pixel-level anomaly segmentation and tracking.

  9. "KAN you hear me?" Exploring Kolmogorov-Arnold Networks for Spoken Language Understanding

    cs.CL 2025-05 conditional novelty 5.0 of 10

    Placing a KAN layer between two linear layers improves spoken language understanding accuracy over linear-only baselines on several speech-intent datasets.

  10. MatrixKAN: Parallelized Kolmogorov-Arnold Network

    cs.LG 2025-02 conditional novelty 5.0 of 10

    MatrixKAN replaces KAN's recursive B-spline evaluation with precomputed matrix multiplications, making training time nearly independent of spline degree.

  11. Low Tensor-Rank Adaptation of Kolmogorov--Arnold Networks

    cs.LG 2025-02 conditional novelty 5.0 of 10

    A low tensor-rank adaptation (LoTRA) method and learning-rate guidance enable efficient fine-tuning of Kolmogorov-Arnold networks, validated on PDE solving and representation tasks.

  12. TimeKAN: KAN-based Frequency Decomposition Learning Architecture for Long-term Time Series Forecasting

    cs.LG 2025-02 conditional novelty 5.0 of 10

    A frequency-decomposing KAN architecture achieves state-of-the-art or near-state-of-the-art long-term forecasting on five of six datasets with 12-38K parameters.

  13. FORTRESS: Function-composition Optimized Real-Time Resilient Structural Segmentation via Kolmogorov-Arnold Enhanced Spatial Attention Networks

    cs.CV 2025-07 reject novelty 4.0 of 10

    FORTRESS combines depthwise separable convolutions and a gated Kolmogorov-Arnold module to report F1 of 0.771 and mIoU of 0.677 on the CSDD benchmark, but the core KAN contribution is not isolated by ablation.

  14. Free-Knots Kolmogorov-Arnold Network: On the Analysis of Spline Knots and Advancing Stability

    cs.LG 2025-01 reject novelty 4.0 of 10

    A free-knot variant of Kolmogorov-Arnold networks reports higher accuracy with fewer parameters than fixed-grid KAN, but its central smoothing regularizer is mathematically inert and the knot bound proof is not rigorous.

  15. On Computational Limits and Provably Efficient Criteria of Visual Autoregressive Models: A Fine-Grained Complexity Analysis

    cs.LG 2025-01 reject novelty 4.0 of 10

    Under SETH, the paper claims VAR models cannot be approximated faster than O(n^4) when attention entries are Theta(sqrt(log n)), but can be approximated in O(n^{2+o(1)}) when entries are o(sqrt(log n)).

  16. Circuit Complexity Bounds for Visual Autoregressive Model

    stat.ML 2025-01 reject novelty 4.0 of 10

    The authors show that a simplified formalization of the VAR image generation model lies in DLOGTIME-uniform TC0, meaning it can be simulated by constant-depth threshold circuits with polynomial size and precision.

  17. KM-UNet KAN Mamba UNet for medical image segmentation

    eess.IV 2025-01 conditional novelty 4.0 of 10

    KM-UNet reports the best average IoU and F1 among compared methods on five medical segmentation benchmarks with only 7.35M parameters.

  18. KAE: Kolmogorov-Arnold Auto-Encoder for Representation Learning

    cs.LG 2024-12 conditional novelty 4.0 of 10

    KAE, an autoencoder with polynomial KAN layers, achieves lower reconstruction error and better downstream task performance than standard autoencoders and other KAN variants on four image benchmarks.

Pith tools