mlsanityData to 14 Oct 2026

MoST: Efficient Monarch Sparse Tuning for 3D Representation Learning

arXiv 2503.18368 ↗First posted 24 Mar 2025cs.CV cs.LGcited by 6

Not on the map. Why?

Abstract (arXiv)

We introduce Monarch Sparse Tuning (MoST), the first reparameterization-based parameter-efficient fine-tuning (PEFT) method tailored for 3D representation learning. Unlike existing adapter-based and prompt-tuning 3D PEFT methods, MoST introduces no additional inference overhead and is compatible with many 3D representation learning backbones. At its core, we present a new family of structured matrices for 3D point clouds, Point Monarch, which can capture local geometric features of irregular points while offering high expressiveness. MoST reparameterizes the dense update weight matrices as our sparse Point Monarch matrices, significantly reducing parameters while retaining strong performance. Experiments on various backbones show that MoST is simple, effective, and highly generalizable. It captures local features in point clouds, achieving state-of-the-art results on multiple benchmarks, e.g., 97.5% acc. on ScanObjectNN (PB_50_RS) and 96.2% on ModelNet40 classification, while it can also combine with other matrix decompositions (e.g., Low-rank, Kronecker) to further reduce parameters.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 57 of its 67 references that are in the map. First posted on arXiv 24 Mar 2025.

Builds on

57 of 67 references.

References · 57

  1. 2017Attention Is All You Needcited by 60,361
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  3. 2020Language Models are Few-Shot Learnerscited by 32,419
  4. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  5. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,403
  6. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,248
  7. 2023Mamba: Linear-Time Sequence Modeling with Selective State Spacescited by 4,718
  8. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,401
  9. 2019Parameter-Efficient Transfer Learning for NLPcited by 4,236
  10. 2021Prefix-Tuning: Optimizing Continuous Prompts for Generationcited by 3,915
  11. 2021The Power of Scale for Parameter-Efficient Prompt Tuningcited by 3,795
  12. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,897
38 more
  1. 2015Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Codingcited by 2,329
  2. 2018The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networkscited by 2,262
  3. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,223
  4. 2022Visual Prompt Tuningcited by 1,681
  5. 2021BitFit: Simple Parameter-efficient Fine-tuning for Transformer-based Masked Language-modelscited by 1,245
  6. 2019KPConv: Flexible and Deformable Convolution for Point Cloudscited by 1,205
  7. 2017ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devicescited by 1,200
  8. 2018PointRCNN: 3D Object Proposal Generation and Detection from Point Cloudcited by 941
  9. 2018PointCNN: Convolution On X-Transformed Pointscited by 764
  10. 2020PCT: Point cloud transformercited by 732
  11. 2022AdaptFormer: Adapting Vision Transformers for Scalable Visual Recognitioncited by 719
  12. 2021Point-BERT: Pre-training 3D Point Cloud Transformers with Masked Point Modelingcited by 653
  13. 2019Deep Learning for 3D Point Clouds: A Surveycited by 558
  14. 2019Revisiting Point Cloud Classification: A New Benchmark Dataset and Classification Model on Real-World Datacited by 532
  15. 2022Masked Autoencoders for Point Cloud Self-supervised Learningcited by 459
  16. 2022PointNeXt: Revisiting PointNet++ with Improved Training and Scaling Strategiescited by 436
  17. 2022Scaling & Shifting Your Features: A New Baseline for Efficient Model Tuningcited by 294
  18. 2022Point-M2AE: Multi-scale Masked Autoencoders for Hierarchical Point Cloud Pre-trainingcited by 264
  19. 2023Scaling Down to Scale Up: A Guide to Parameter-Efficient Fine-Tuningcited by 173
  20. 2024PointMamba: A Simple State Space Model for Point Cloud Analysiscited by 167
  21. 2023Contrast with Reconstruct: Contrastive 3D Representation Learning Guided by Generative Pretrainingcited by 164
  22. 2021Offboard 3D Object Detection from Point Cloud Sequencescited by 145
  23. 2022Learning 3D Representations from 2D Pre-trained Models via Image-to-Point Masked Autoencoderscited by 144
  24. 2024ShapeLLM: Universal 3D Object Understanding for Embodied Interactioncited by 137
  25. 2022FacT: Factor-Tuning for Lightweight Adaptation on Vision Transformercited by 133
  26. 2022Autoencoders as Cross-Modal Teachers: Can Pretrained 2D Image Transformers Help 3D Representation Learning?cited by 122
  27. 2023PointGPT: Auto-regressively Generative Pre-training from Point Cloudscited by 115
  28. 2022Monarch: Expressive Structured Matrices for Efficient and Accurate Trainingcited by 106
  29. 2019Learning Fast Algorithms for Linear Transforms Using Butterfly Factorizationscited by 76
  30. 2024Parameter-Efficient Fine-Tuning with Discrete Fourier Transformcited by 71
  31. 2023Instance-aware Dynamic Prompt Tuning for Pre-trained Point Cloud Modelscited by 68
  32. 2020Self-Supervised Few-Shot Learning on Point Cloudscited by 66
  33. 2024Mamba3D: Enhancing Local Features for 3D Point Cloud Analysis via State Space Modelcited by 57
  34. 2022Unsupervised Point Cloud Representation Learning with Deep Neural Networks: A Surveycited by 54
  35. 2024Dynamic Adapter Meets Prompt Tuning: Parameter-Efficient Transfer Learning for Point Cloud Analysiscited by 53
  36. 2024Point Cloud Matters: Rethinking the Impact of Different Observation Spaces on Robot Learningcited by 45
  37. 2023Revisiting the Parameter Efficiency of Adapters from the Perspective of Precision Redundancycited by 44
  38. 2024Parameter-Efficient Fine-Tuning in Spectral Domain for Point Cloud Learningcited by 43

50 most cited of 57 All 57 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 6.

6 papers.

  1. 2026Mantis: Mamba-native Tuning is Efficient for 3D Point Cloud Foundation Modelscited by 6
  2. 2026Parameter-Efficient Fine-Tuning for Spiking Point Cloud Modelscited by 3
  3. 2026An Adapter-free Fine-tuning Approach for Tuning 3D Foundation Modelscited by 1
  4. 2026Partition-Invariant Tuning for 3D Scene Understandingcited by 0
  5. 2026Position Anchor Tuning: Towards Efficient Adaptation of Pre-Trained Point Cloud Transformerscited by 0
  6. 2026GAPrompt++: Multi-Granular Geometry-Aware Point Cloud Prompt for 3D Vision Modelcited by 0