Pith. sign in

REVIEW 3 major objections 4 minor 63 references

PMA: Towards Parameter-Efficient Point Cloud Understanding via Point Mamba Adapter

T0 review · 3 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read Fusing all layers of a frozen pre-trained point-cloud model with a Mamba adapter lets parameter-efficient fine-tuning beat full fine-tuning while using 99% fewer trainable parameters.

desk verdict Useful parameter-efficient adapter for point clouds, but the headline ordering mechanism is not actually optimized end-to-end the way the paper claims. read the letter →

arxiv 2505.20941 v1 pith:B2IGCX6U submitted 2025-05-27 cs.CV

Yaohua Zha , Yanzi Wang , Hang Guo , Jinpeng Wang , Tao Dai , Bin Chen , Zhihao Ouyang , Xue Yuerong
show 2 more authors
Ke Chen Shu-Tao Xia
This is my paper
classification cs.CV
keywords pointcloudunderstandingparameter-efficientfine-tuningMambastatespacemodelsintermediatefeatures3Dperceptionfeaturefusion
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to show that a frozen pre-trained point-cloud model holds usable information in every layer, not just the last one, and that a lightweight Mamba-based adapter can harvest it. The proposed Point Mamba Adapter (PMA) concatenates tokens from all layers into one ordered sequence, reorders them with a geometry-constrained gate prompt generator (G2PG), and fuses them with a state-space model. Across ScanObjectNN, ModelNet40, ShapeNetPart, and few-shot benchmarks, PMA is reported to beat full fine-tuning and prior parameter-efficient methods while training only 4.9M parameters on the 360M-parameter PointGPT-L backbone. If true, this makes large pre-trained 3D models much cheaper to adapt, especially for fine-grained tasks like part segmentation.

What carries the argument

The central object is the Point Mamba Adapter, built on Mamba, a selective state-space sequence model whose recurrence is $h_t = A h_{t-1} + B x_t$ with output $y_t = (C+P)h_t + D x_t$. The adapter fuses the $L \times M$ patch tokens collected from all layers of the frozen backbone. The second load-bearing object is the geometry-constrained gate prompt generator (G2PG), which builds a KNN geometric graph over tokens, aggregates neighbor features with down-projection and max pooling, applies softmax, and then produces two things: geometric prompts $P$ added to the output matrix $C$, and a one-hot plus argmax index that reorders tokens before fusion. The paper claims this injects spatial-neighborhood constraints into the sequence model and makes the ordering task-adaptable.

What would settle it

Inspect the released code for the reorder branch: if the argmax-selected indices are detached or no straight-through estimator is used, then the ordering is not optimized by gradient descent; retraining on PB-T50-RS with the learned ordering replaced by a fixed random permutation would then show whether the reported 86.43% accuracy actually depends on the dynamic order.

Watch

Extended reading notes

Core claim

The central claim is that intermediate features from a frozen pre-trained backbone are nearly as informative as the final features and contain complementary information, so discarding them wastes most of the model's representational power. PMA is the proposed fix: extract patch-token outputs from every Transformer layer, feed them through a shared G2PG that produces both a geometric prompt for Mamba's output matrix and a dynamically learned token order, concatenate all layers' tokens chronologically, and run a Mamba adapter whose output state is modulated by the geometric prompt. The fused output is combined with the last-layer CLS token and final features before the task head. The paper reports that this arrangement beats full fine-tuning and earlier PEFT methods across classification, part segmentation, and few-shot learning, with a 99% reduction in trainable parameters on PointGPT-L, and that the dynamic geometry-constrained ordering outperforms fixed Hilbert, z-order, and axis-orderings.

Load-bearing premise

The learned order is computed by a hard “take the maximum” step over one-hot codes, and the paper does not say how gradients pass through it; if they do not, the ordering is not truly trained end-to-end and the reported benefit of dynamic ordering over fixed Hilbert or z-order curves is not established.

Editorial extensions

If this is right

  • If PMA's claims hold, parameter-efficient fine-tuning can close most of the gap to full fine-tuning on fine-grained point-level tasks like part segmentation.
  • The 360M-parameter PointGPT-L backbone can be adapted with only 4.9M trainable parameters while improving classification accuracy on ScanObjectNN and ModelNet40.
  • Point-cloud PEFT methods that use only final-layer features leave measurable performance unused; multi-layer fusion becomes a standard component of future adapters.
  • The learned geometry-constrained ordering gives a reported gain over Hilbert, z-order, and axis-orderings (86.43 versus about 85.9 on the PB-T50-RS variant), so ordering strategy matters for sequence-based fusion.
  • PMA transfers across different pre-trained backbones, including Point-BERT, Point-MAE, Recon, and PointGPT-L, suggesting the approach is not tied to one pre-training method.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: because one-hot plus argmax has no gradient, the claim of end-to-end learned ordering depends on an unstated gradient approximation such as a straight-through estimator; checking the released code for that estimator is the fastest way to test the paper's main novelty.
  • Editorial inference: the concatenate-all-layers-and-scan design is not specific to point clouds, so the same adapter recipe could be transferred to other frozen architectures with layer-wise patch tokens, such as 2D or video transformers.
  • Editorial inference: the sequence length grows as $L \times M$, so a practical deployment of PMA should be measured for latency and memory on long point clouds; the linear-time advantage of Mamba may shrink when the fused sequence is very long.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. This paper proposes Point Mamba Adapter (PMA), a parameter-efficient fine-tuning method for point cloud Transformers. PMA freezes the pre-trained backbone, collects token features from every layer, concatenates them into a long token sequence, and processes the sequence with a Mamba state-space model. A shared Geometry-constrained Gate Prompt Generator (G2PG) produces per-layer geometric prompts injected into Mamba's output matrix C and generates a token ordering via Softmax, one-hot encoding, and argmax, with the goal of dynamic, task-specific spatial ordering. The method is evaluated on classification (ScanObjectNN, ModelNet40), part segmentation (ShapeNetPart), and few-shot classification, using Point-BERT, Point-MAE, Recon, and PointGPT-L as frozen backbones, and reports state-of-the-art PEFT accuracy with 1 to 5 million trainable parameters. Code is publicly released.

Significance. The core idea of fusing intermediate features of a frozen backbone with a linear-complexity sequence model is sensible and orthogonal to existing point-cloud PEFT methods. The paper's strengths are its breadth of experiments (three tasks, four backbones, comparisons to full fine-tuning and several PEFT baselines), the public code release, and the motivating observation experiment in Sec. 4.1. If the dynamic ordering were genuinely optimized end-to-end, the G2PG mechanism would be a useful contribution. However, the current description does not support that claim, and the reported baseline numbers are internally inconsistent, so the central empirical claims need substantiation before the paper can be accepted.

major comments (3)
  1. [Sec. 4.4.2, Fig. 3, Sec. 5.5.2] The token reordering is described as Softmax followed by One-hot encoding and Argmax. Argmax is non-differentiable almost everywhere, and because the backbone is frozen, gradients from the task loss cannot reach the G2PG through the reorder path in Eqs. (7)-(8); the only gradient path is through the geometric prompt P_i. Therefore the ordering is not actually optimized by gradient descent for the downstream task, contrary to the claims of "end-to-end" optimization in Sec. 5.5.1 and Sec. 5.5.2. As a result, the 0.48-point advantage of "Ours" over Z-Order in Table 7 cannot be attributed to task-specific ordering optimization; it may come entirely from the prompt generation branch. Please specify a gradient estimator (e.g., straight-through estimator, Gumbel-Softmax, or differentiable sorting) or provide a controlled experiment that isolates the ordering effect from the prompt effect.
  2. [Table 1 vs. Table 2] The baseline numbers are internally inconsistent. Table 1 lists "Point-MAE w/ IDPT" with 2.7M trainable parameters and accuracies 95.18/93.29/90.63 on the three ScanObjectNN variants, values identical to the Recon FFT row, whereas Table 2 lists Point-MAE w/ IDPT with 1.7M parameters and accuracies 91.22/90.02/84.94. Similarly, PointGPT-L w/ PointGST is given as 0.6M parameters in Table 1 but 2.4M in Table 2, with the same accuracy values. Because IDPT, DAPT, and PointGST are the main PEFT baselines, these discrepancies directly affect the claimed rankings and the parameter-efficiency comparison. Please correct and reconcile all baseline rows, including #TP and accuracy, across Tables 1, 2, 3, and 5.
  3. [Tables 6 and 7 (and main accuracy tables)] The component ablation in Table 6 and the ordering-strategy comparison in Table 7 report single-run accuracies, with gaps as small as 0.48 points (86.43 vs. 85.95). The main classification and segmentation tables also report no variance or number of seeds. Given that few-shot results in Table 4 show standard deviations of 1-5 points, the small margins in Tables 1, 2, 3, and 7 may be within run-to-run noise. Please report mean and standard deviation over at least three seeds for the central comparisons, or explicitly state the experimental protocol used to obtain the reported numbers.
minor comments (4)
  1. [Preliminaries and Sec. 4.4.2] The dimensions of A, B, C, and D are stated inconsistently: Eqs. (1)-(4) define A in R^{S x S}, B in R^{1 x N}, C in R^{1 x N}, and D in R^{1 x N}, but Sec. 4.4.2 says the dimension of C is set to the number of patches, with S=128 as an example. Please clarify whether S denotes the Mamba state dimension or the number of tokens, and align the notation throughout.
  2. [Sec. 4.1, Fig. 1] The observation experiment in Sec. 4.1 cites drops of 2.6% and 0.14% when using the first 3 layers, but Fig. 1 does not show absolute accuracy values or error bars, making it difficult to assess the significance of these drops. Please include numerical values and, if possible, variance in the figure or its caption.
  3. [Eq. (9)] Equation (9) refers to features F_pre from the first N-1 layers, but the architecture description in Sec. 4.3 uses L layers and never defines N. Please replace N with L or define N explicitly.
  4. [References] Some references are incomplete or inconsistently formatted, e.g., PointGPT [2] gives only "2024" without a venue, and several arXiv preprints do not include arXiv identifiers. Please normalize the reference list.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: PMA is an empirical architecture study whose claims are tested on held-out data, and the one-hot/argmax ordering caveat is a technical correctness concern, not a circular step.

full rationale

The paper is an empirical architecture study rather than a derivation chain. Its central claim, that PMA outperforms full fine-tuning and existing PEFT methods, is supported by held-out evaluations on ScanObjectNN, ModelNet40, ShapeNetPart, and ModelNet40 few-shot benchmarks against external baselines, so no result is defined in terms of its own fitted inputs. The observation experiment in Sec. 4.1 motivates the multi-layer fusion design but does not set the target performance metric, and the Mamba Adapter and G2PG are trained with a standard downstream task loss. Self-citations such as [51], [52], [53], and [54] appear only as prior works and baselines, not as load-bearing uniqueness theorems or as substitutes for validation. No equation reduces to its own input, and no fitted parameter is renamed as a prediction. A genuine technical caveat exists in Sec. 4.4.2 and Figure 3: the token ordering is generated via Softmax, One-hot, then Argmax, which is non-differentiable, so the paper's claim of end-to-end optimization of the sorting process is not substantiated as written. This is a correctness or reproducibility concern, not circularity, and it does not change the circularity score.

Assumptions & free parameters 1 free parameters · 3 assumptions · 0 invented entities

The central claim is purely empirical; no free parameters are fitted in a derivation sense, though architecture hyperparameters like the state dimension are chosen. The load-bearing assumptions are about the usefulness of intermediate features, the effectiveness of Mamba for this fusion, and the trainability of the discrete reordering.

free parameters (1)
  • Mamba state dimension S = 128 for classification on ScanObjectNN
    Set equal to the number of point patches; a chosen hyperparameter rather than a value fitted to maximize accuracy.
assumptions (3)
  • domain assumption Intermediate features of frozen layers contain complementary information useful for downstream tasks
    Supported by Section 4.1 observation, but measured in a single experiment without error bars or significance testing.
  • domain assumption Mamba's selective state space model can effectively fuse long sequences of layer tokens
    Assumed from prior SSM successes; no complexity or capacity analysis specific to this fused sequence is given.
  • ad hoc to paper The discrete argmax ordering can be optimized with gradient descent
    The paper claims end-to-end dynamic ordering but does not describe a straight-through estimator or other differentiable relaxation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of PMA: Towards Parameter-Efficient Point Cloud Understanding via Point Mamba Adapter." pith.science (2026). https://pith.science/paper/B2IGCX6U

@misc{pith2026250520941,
  author       = {Pith},
  title        = {Pith review of: PMA: Towards Parameter-Efficient Point Cloud Understanding via Point Mamba Adapter},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/B2IGCX6U}},
  note         = {Machine review of arXiv:2505.20941}
}
read the original abstract

Applying pre-trained models to assist point cloud understanding has recently become a mainstream paradigm in 3D perception. However, existing application strategies are straightforward, utilizing only the final output of the pre-trained model for various task heads. It neglects the rich complementary information in the intermediate layer, thereby failing to fully unlock the potential of pre-trained models. To overcome this limitation, we propose an orthogonal solution: Point Mamba Adapter (PMA), which constructs an ordered feature sequence from all layers of the pre-trained model and leverages Mamba to fuse all complementary semantics, thereby promoting comprehensive point cloud understanding. Constructing this ordered sequence is non-trivial due to the inherent isotropy of 3D space. Therefore, we further propose a geometry-constrained gate prompt generator (G2PG) shared across different layers, which applies shared geometric constraints to the output gates of the Mamba and dynamically optimizes the spatial order, thus enabling more effective integration of multi-layer information. Extensive experiments conducted on challenging point cloud datasets across various tasks demonstrate that our PMA elevates the capability for point cloud understanding to a new level by fusing diverse complementary intermediate features. Code is available at https://github.com/zyh16143998882/PMA.

Figures

Figures reproduced from arXiv: 2505.20941 by the authors.

Figure 1
Figure 1. The impact of freezing different numbers of pre-trained [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. The parameter-efficient fine-tuning pipeline based on our Point Mamba Adapter. It consists of three main components: the [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Details of our Geometry-constrained Gate Prompt Gen [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

63 extracted references · 44 canonical work pages

  1. [1]

    Shapenet: An information-rich 3d model repository.arXiv preprint arXiv:1512.03012, 2015

    Angel X Chang, Thomas Funkhouser, Leonidas Guibas, Pat Hanrahan, Qixing Huang, Zimo Li, Silvio Savarese, Manolis Savva, Shuran Song, Hao Su, et al. Shapenet: An information-rich 3d model repository.arXiv preprint arXiv:1512.03012, 2015. 3

    arXiv 2015

  2. [2]

    Pointgpt: Auto-regressively generative pre- training from point clouds

    Guangyan Chen, Meiling Wang, Yi Yang, Kai Yu, Li Yuan, and Yufeng Yue. Pointgpt: Auto-regressively generative pre- training from point clouds. 2024. 1, 6, 7

    work page 2024

  3. [3]

    Adaptformer: Adapt- ing vision transformers for scalable visual recognition

    Shoufa Chen, Chongjian Ge, Zhan Tong, Jiangliu Wang, Yibing Song, Jue Wang, and Ping Luo. Adaptformer: Adapt- ing vision transformers for scalable visual recognition. In Advances in Neural Information Processing Systems, 2022. 2, 8

    work page 2022

  4. [4]

    Delta tuning: A comprehen- sive study of parameter efficient methods for pre-trained lan- guage models.arXiv preprint arXiv:2203.06904, 2022

    Ning Ding, Yujia Qin, Guang Yang, Fuchao Wei, Zong- han Yang, Yusheng Su, Shengding Hu, Yulin Chen, Chi- Min Chan, Weize Chen, et al. Delta tuning: A comprehen- sive study of parameter efficient methods for pre-trained lan- guage models.arXiv preprint arXiv:2203.06904, 2022

    arXiv 2022

  5. [5]

    Parameter-efficient fine-tuning of large-scale pre-trained language models.Nature Machine In- telligence, 5(3):220–235, 2023

    Ning Ding, Yujia Qin, Guang Yang, Fuchao Wei, Zonghan Yang, Yusheng Su, Shengding Hu, Yulin Chen, Chi-Min Chan, Weize Chen, et al. Parameter-efficient fine-tuning of large-scale pre-trained language models.Nature Machine In- telligence, 5(3):220–235, 2023. 2

    work page 2023

  6. [6]

    Autoencoders as cross-modal teachers: Can pretrained 2d image transform- ers help 3d representation learning? Kigali, Rwanda, 2023

    Runpei Dong, Zekun Qi, Linfeng Zhang, Junbo Zhang, Jian- jian Sun, Zheng Ge, Li Yi, and Kaisheng Ma. Autoencoders as cross-modal teachers: Can pretrained 2d image transform- ers help 3d representation learning? Kigali, Rwanda, 2023. 6, 7

    work page 2023

  7. [7]

    Shape2Scene: 3D Scene Representation Learning Through Pre-training on Shape Data

    Tuo Feng, Wenguan Wang, Ruijie Quan, and Yi Yang. Shape2scene: 3d scene representation learning through pre- training on shape data.arXiv preprint arXiv:2407.10200,

    work page Pith review arXiv

  8. [8]

    Mamba: Linear-time sequence modeling with selective state spaces.arXiv preprint arXiv:2312.00752, 2023

    Albert Gu and Tri Dao. Mamba: Linear-time sequence modeling with selective state spaces.arXiv preprint arXiv:2312.00752, 2023. 2, 3

    arXiv 2023

Show all 63 references
  1. [9]

    Hippo: Recurrent memory with optimal polynomial projections.Advances in neural information processing sys- tems, 33:1474–1487, 2020

    Albert Gu, Tri Dao, Stefano Ermon, Atri Rudra, and Christo- pher R´e. Hippo: Recurrent memory with optimal polynomial projections.Advances in neural information processing sys- tems, 33:1474–1487, 2020. 2

    2020

  2. [10]

    Efficiently modeling long sequences with structured state spaces.arXiv preprint arXiv:2111.00396, 2021

    Albert Gu, Karan Goel, and Christopher R ´e. Efficiently modeling long sequences with structured state spaces.arXiv preprint arXiv:2111.00396, 2021. 2, 3

    2021 arXiv

  3. [11]

    Combining recurrent, convolutional, and continuous-time models with linear state space layers.Advances in neural information processing sys- tems, 34:572–585, 2021

    Albert Gu, Isys Johnson, Karan Goel, Khaled Saab, Tri Dao, Atri Rudra, and Christopher R ´e. Combining recurrent, convolutional, and continuous-time models with linear state space layers.Advances in neural information processing sys- tems, 34:572–585, 2021

    2021

  4. [12]

    On the parameterization and initialization of diagonal state space models.Advances in Neural Information Processing Systems, 35:35971–35983, 2022

    Albert Gu, Karan Goel, Ankit Gupta, and Christopher R ´e. On the parameterization and initialization of diagonal state space models.Advances in Neural Information Processing Systems, 35:35971–35983, 2022. 2

    2022

  5. [13]

    Adaptir: Parameter efficient multi-task adaptation for pre-trained image restoration models.arXiv preprint arXiv:2312.08881, 2023

    Hang Guo, Tao Dai, Yuanchao Bai, Bin Chen, Shu-Tao Xia, and Zexuan Zhu. Adaptir: Parameter efficient multi-task adaptation for pre-trained image restoration models.arXiv preprint arXiv:2312.08881, 2023. 2

    2023 arXiv

  6. [14]

    Mambairv2: Attentive state space restoration.arXiv preprint arXiv:2411.15269,

    Hang Guo, Yong Guo, Yaohua Zha, Yulun Zhang, Wenbo Li, Tao Dai, Shu-Tao Xia, and Yawei Li. Mambairv2: Attentive state space restoration.arXiv preprint arXiv:2411.15269,

    arXiv

  7. [15]

    Mambair: A simple baseline for image restoration with state-space model.arXiv preprint arXiv:2402.15648, 2024

    Hang Guo, Jinmin Li, Tao Dai, Zhihao Ouyang, Xudong Ren, and Shu-Tao Xia. Mambair: A simple baseline for image restoration with state-space model.arXiv preprint arXiv:2402.15648, 2024. 2

    2024 arXiv

  8. [16]

    Demystify mamba in vision: A linear attention perspective.arXiv preprint arXiv:2405.16605, 2024

    Dongchen Han, Ziyi Wang, Zhuofan Xia, Yizeng Han, Yi- fan Pu, Chunjiang Ge, Jun Song, Shiji Song, Bo Zheng, and Gao Huang. Demystify mamba in vision: A linear attention perspective.arXiv preprint arXiv:2405.16605, 2024. 5

    2024 arXiv

  9. [17]

    Mamba3d: Enhancing local features for 3d point cloud anal- ysis via state space model.arXiv preprint arXiv:2404.14966,

    Xu Han, Yuan Tang, Zhaoxuan Wang, and Xianzhi Li. Mamba3d: Enhancing local features for 3d point cloud anal- ysis via state space model.arXiv preprint arXiv:2404.14966,

    arXiv

  10. [18]

    David Hilbert and David Hilbert. ¨Uber die stetige abbildung einer linie auf ein fl ¨achenst¨uck.Dritter Band: Analysis· Grundlagen der Mathematik· Physik Verschiedenes: Nebst Einer Lebensgeschichte, pages 1–2, 1935. 8

    1935

  11. [19]

    Parameter-efficient transfer learning for nlp

    Neil Houlsby, Andrei Giurgiu, Stanislaw Jastrzebski, Bruna Morrone, Quentin De Laroussilhe, Andrea Gesmundo, Mona Attariyan, and Sylvain Gelly. Parameter-efficient transfer learning for nlp. InInternational Conference on Machine Learning, pages 2790–2799. PMLR, 2019. 2, 8

    2019

  12. [20]

    Lora: Low-rank adaptation of large language models.arXiv preprint arXiv:2106.09685, 2021

    Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen- Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. Lora: Low-rank adaptation of large language models.arXiv preprint arXiv:2106.09685, 2021. 2, 8

    2021 arXiv

  13. [21]

    Vi- sual prompt tuning

    Menglin Jia, Luming Tang, Bor-Chun Chen, Claire Cardie, Serge Belongie, Bharath Hariharan, and Ser-Nam Lim. Vi- sual prompt tuning. InProceedings of the European Confer- ence on Computer Vision (ECCV), pages 709–727, Tel Aviv, Israel, 2022. 2, 8

    2022

  14. [22]

    Revisiting the parameter efficiency of adapters from the perspective of precision redundancy

    Shibo Jie, Haoqing Wang, and Zhi-Hong Deng. Revisiting the parameter efficiency of adapters from the perspective of precision redundancy. InProceedings of the IEEE/CVF In- ternational Conference on Computer Vision, pages 17217– 17226, 2023. 8

    2023

  15. [23]

    The power of scale for parameter-efficient prompt tuning.arXiv preprint arXiv:2104.08691, 2021

    Brian Lester, Rami Al-Rfou, and Noah Constant. The power of scale for parameter-efficient prompt tuning.arXiv preprint arXiv:2104.08691, 2021. 2

    2021 arXiv

  16. [24]

    Parameter-efficient fine-tuning in spectral domain for point cloud learning.arXiv preprint arXiv:2410.08114, 2024

    Dingkang Liang, Tianrui Feng, Xin Zhou, Yumeng Zhang, Zhikang Zou, and Xiang Bai. Parameter-efficient fine-tuning in spectral domain for point cloud learning.arXiv preprint arXiv:2410.08114, 2024. 2, 6, 7

    2024 arXiv

  17. [25]

    Point- mamba: A simple state space model for point cloud analysis

    Dingkang Liang, Xin Zhou, Xinyu Wang, Xingkui Zhu, Wei Xu, Zhikang Zou, Xiaoqing Ye, and Xiang Bai. Point- mamba: A simple state space model for point cloud analysis. arXiv preprint arXiv:2402.10739, 2024. 1, 2, 6

    2024 arXiv

  18. [26]

    Masked discrimi- nation for self-supervised learning on point clouds

    Haotian Liu, Mu Cai, and Yong Jae Lee. Masked discrimi- nation for self-supervised learning on point clouds. InPro- ceedings of the European Conference on Computer Vision (ECCV), pages 657–675, Tel Aviv, Israel, 2022. 7

    2022

  19. [27]

    P-tuning v2: Prompt tuning can be com- parable to fine-tuning universally across scales and tasks

    Xiao Liu, Kaixuan Ji, Yicheng Fu, Zhengxiao Du, Zhilin Yang, and Jie Tang. P-tuning v2: Prompt tuning can be com- parable to fine-tuning universally across scales and tasks. arXiv preprint arXiv:2110.07602, 2021. 2

    2021 arXiv

  20. [28]

    Vmamba: Visual state space model.arXiv preprint arXiv:2401.10166,

    Yue Liu, Yunjie Tian, Yuzhong Zhao, Hongtian Yu, Lingxi Xie, Yaowei Wang, Qixiang Ye, and Yunfan Liu. Vmamba: Visual state space model.arXiv preprint arXiv:2401.10166,

    arXiv

  21. [29]

    Re- thinking network design and local geometry in point cloud: A simple residual mlp framework

    Xu Ma, Can Qin, Haoxuan You, Haoxi Ran, and Yun Fu. Re- thinking network design and local geometry in point cloud: A simple residual mlp framework. InProceedings of Inter- national Conference on Learning Representations (ICLR), page 31, Online, 2022. 1, 6

    2022

  22. [30]

    An end-to- end transformer model for 3d object detection

    Ishan Misra, Rohit Girdhar, and Armand Joulin. An end-to- end transformer model for 3d object detection. InProceed- ings of the IEEE/CVF International Conference on Com- puter Vision, pages 2906–2917, 2021. 1

    2021

  23. [31]

    A computer oriented geodetic data base and a new technique in file sequencing

    Guy M Morton. A computer oriented geodetic data base and a new technique in file sequencing. 1966. 8

    1966

  24. [32]

    Masked autoencoders for point cloud self-supervised learning

    Yatian Pang, Wenxiao Wang, Francis EH Tay, Wei Liu, Yonghong Tian, and Li Yuan. Masked autoencoders for point cloud self-supervised learning. InProceedings of the Euro- pean Conference on Computer Vision (ECCV), Tel Aviv, Is- rael, 2022. 1, 3, 6, 7

    2022

  25. [33]

    Pointnet: Deep learning on point sets for 3d classification and segmentation

    Charles R Qi, Hao Su, Kaichun Mo, and Leonidas J Guibas. Pointnet: Deep learning on point sets for 3d classification and segmentation. InProceedings of the IEEE/CVF Confer- ence on Computer Vision and Pattern Recognition (CVPR), pages 652–660, Honolulu, HI, USA, 2017. 1, 7

    2017

  26. [34]

    Pointnet++: Deep hierarchical feature learning on point sets in a metric space

    Charles Ruizhongtai Qi, Li Yi, Hao Su, and Leonidas J Guibas. Pointnet++: Deep hierarchical feature learning on point sets in a metric space. InProceedings of Advances in Neural Information Processing Systems (NeurIPS), page 30, Long Beach, CA, USA, 2017. 1, 6, 7

    2017

  27. [35]

    Contrast with reconstruct: Contrastive 3d representation learning guided by generative pretraining

    Zekun Qi, Runpei Dong, Guofan Fan, Zheng Ge, Xiangyu Zhang, Kaisheng Ma, and Li Yi. Contrast with reconstruct: Contrastive 3d representation learning guided by generative pretraining. InInternational Conference on Machine Learn- ing, 2023. 1, 6, 7

    2023

  28. [36]

    Simplified state space layers for sequence modeling

    Jimmy TH Smith, Andrew Warrington, and Scott W Linder- man. Simplified state space layers for sequence modeling. arXiv preprint arXiv:2208.04933, 2022. 2

    2022 arXiv

  29. [37]

    Parameter-efficient prompt learning for 3d point cloud understanding.arXiv preprint arXiv:2402.15823,

    Hongyu Sun, Yongcai Wang, Wang Chen, Haoran Deng, and Deying Li. Parameter-efficient prompt learning for 3d point cloud understanding.arXiv preprint arXiv:2402.15823,

    arXiv

  30. [38]

    X-3d: Explicit 3d structure modeling for point cloud recog- nition

    Shuofeng Sun, Yongming Rao, Jiwen Lu, and Haibin Yan. X-3d: Explicit 3d structure modeling for point cloud recog- nition. InProceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition, pages 5074–5083,

  31. [39]

    Point- peft: Parameter-efficient fine-tuning for 3d pre-trained mod- els

    Yiwen Tang, Ray Zhang, Zoey Guo, Xianzheng Ma, Bin Zhao, Zhigang Wang, Dong Wang, and Xuelong Li. Point- peft: Parameter-efficient fine-tuning for 3d pre-trained mod- els. InProceedings of the AAAI Conference on Artificial In- telligence, pages 5171–5179, 2024. 1, 2, 3

    2024

  32. [40]

    Multimodal few-shot learning with frozen language models

    Maria Tsimpoukelli, Jacob L Menick, Serkan Cabi, SM Es- lami, Oriol Vinyals, and Felix Hill. Multimodal few-shot learning with frozen language models. InAdvances in Neu- ral Information Processing Systems, pages 200–212, 2021. 2

    2021

  33. [41]

    Revisiting point cloud classification: A new benchmark dataset and classification model on real-world data

    Mikaela Angelina Uy, Quang-Hieu Pham, Binh-Son Hua, Thanh Nguyen, and Sai-Kit Yeung. Revisiting point cloud classification: A new benchmark dataset and classification model on real-world data. InProceedings of IEEE/CVF In- ternational Conference on Computer Vision (ICCV), page...

    2019

  34. [42]

    Attention is all you need

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszko- reit, Llion Jones, Aidan N Gomez, Lukasz Kaiser, and Illia Polosukhin. Attention is all you need. InProceedings of Ad- vances in Neural Information Processing Systems (NeurIPS), page 30, Long Beach, CA, USA, 2017. 2, 7

    2017

  35. [43]

    Unsupervised point cloud pre-training via oc- clusion completion

    Hanchen Wang, Qi Liu, Xiangyu Yue, Joan Lasenby, and Matt J Kusner. Unsupervised point cloud pre-training via oc- clusion completion. InProceedings of the IEEE/CVF Inter- national Conference on Computer Vision, pages 9782–9792,

  36. [44]

    Dynamic graph cnn for learning on point clouds.Acm Transactions On Graphics (TOG), 38(5):1–12, 2019

    Yue Wang, Yongbin Sun, Ziwei Liu, Sanjay E Sarma, Michael M Bronstein, and Justin M Solomon. Dynamic graph cnn for learning on point clouds.Acm Transactions On Graphics (TOG), 38(5):1–12, 2019. 1, 7

    2019

  37. [45]

    P2p: Tuning pre-trained image models for point cloud analysis with point-to-pixel prompting

    Ziyi Wang, Xumin Yu, Yongming Rao, Jie Zhou, and Ji- wen Lu. P2p: Tuning pre-trained image models for point cloud analysis with point-to-pixel prompting. InProceed- ings of Advances in Neural Information Processing Systems (NeurIPS), New Orleans, Louisiana, USA, 2022. 2, 6

    2022

  38. [46]

    Point transformer v2: Grouped vector atten- tion and partition-based pooling

    Xiaoyang Wu, Yixing Lao, Li Jiang, Xihui Liu, and Heng- shuang Zhao. Point transformer v2: Grouped vector atten- tion and partition-based pooling. InProceedings of Advances in Neural Information Processing Systems (NeurIPS), pages 33330–33342, New Orleans, Louisiana, USA, 2022. 1

    2022

  39. [47]

    Point transformer v3: Simpler, faster, stronger

    Xiaoyang Wu, Li Jiang, Peng-Shuai Wang, Zhijian Liu, Xi- hui Liu, Yu Qiao, Wanli Ouyang, Tong He, and Hengshuang Zhao. Point transformer v3: Simpler, faster, stronger. InPro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, USA, 2024. 1

    2024

  40. [48]

    3d shapenets: A deep representation for volumetric shapes

    Zhirong Wu, Shuran Song, Aditya Khosla, Fisher Yu, Lin- guang Zhang, Xiaoou Tang, and Jianxiong Xiao. 3d shapenets: A deep representation for volumetric shapes. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 1912–1920, 2015. 5

    1912

  41. [49]

    Point-bert: Pre-training 3d point cloud transformers with masked point modeling

    Xumin Yu, Lulu Tang, Yongming Rao, Tiejun Huang, Jie Zhou, and Jiwen Lu. Point-bert: Pre-training 3d point cloud transformers with masked point modeling. InProceedings of the IEEE/CVF Conference on Computer Vision and Pat- tern Recognition (CVPR), pages 19313–19322, New Or- le...

    2022

  42. [50]

    Bitfit: Simple parameter-efficient fine-tuning for transformer-based masked language-models.arXiv preprint arXiv:2106.10199,

    Elad Ben Zaken, Shauli Ravfogel, and Yoav Goldberg. Bitfit: Simple parameter-efficient fine-tuning for transformer-based masked language-models.arXiv preprint arXiv:2106.10199,

  43. [51]

    Sfr: Semantic-aware feature ren- dering of point cloud

    Yaohua Zha, Rongsheng Li, Tao Dai, Jianyu Xiong, Xin Wang, and Shu-Tao Xia. Sfr: Semantic-aware feature ren- dering of point cloud. InICASSP 2023-2023 IEEE Interna- tional Conference on Acoustics, Speech and Signal Process- ing (ICASSP), pages 1–5. IEEE, 2023. 1, 6

    2023

  44. [52]

    Instance-aware dynamic prompt tuning for pre-trained point cloud models

    Yaohua Zha, Jinpeng Wang, Tao Dai, Bin Chen, Zhi Wang, and Shu-Tao Xia. Instance-aware dynamic prompt tuning for pre-trained point cloud models. InProceedings of IEEE/CVF International Conference on Computer Vision (ICCV), pages 14161–14170, Paris, France, 2023. 1, 2, 3, 6, 7, 8

    2023

  45. [53]

    Towards compact 3d representations via point feature enhancement masked au- toencoders

    Yaohua Zha, Huizhen Ji, Jinmin Li, Rongsheng Li, Tao Dai, Bin Chen, Zhi Wang, and Shu-Tao Xia. Towards compact 3d representations via point feature enhancement masked au- toencoders. InProceedings of the AAAI Conference on Arti- ficial Intelligence (AAAI), V ANCOUVER, CANADA, ...

    2024

  46. [54]

    Lcm: Locally constrained compact point cloud model for masked point modeling.arXiv preprint arXiv:2405.17149, 2024

    Yaohua Zha, Naiqi Li, Yanzi Wang, Tao Dai, Hang Guo, Bin Chen, Zhi Wang, Zhihao Ouyang, and Shu-Tao Xia. Lcm: Locally constrained compact point cloud model for masked point modeling.arXiv preprint arXiv:2405.17149, 2024. 1, 2, 6

    2024 arXiv

  47. [55]

    Point- m2ae: Multi-scale masked autoencoders for hierarchical point cloud pre-training

    Renrui Zhang, Ziyu Guo, Peng Gao, Rongyao Fang, Bin Zhao, Dong Wang, Yu Qiao, and Hongsheng Li. Point- m2ae: Multi-scale masked autoencoders for hierarchical point cloud pre-training. InProceedings of Advances in Neural Information Processing Systems (NeurIPS), New Or- leans, ...

    2022

  48. [56]

    Tip- adapter: Training-free adaption of clip for few-shot classifi- cation

    Renrui Zhang, Wei Zhang, Rongyao Fang, Peng Gao, Kun- chang Li, Jifeng Dai, Yu Qiao, and Hongsheng Li. Tip- adapter: Training-free adaption of clip for few-shot classifi- cation. InProceedings of the European Conference on Com- puter Vision, page 493–510, 2022. 2

    2022

  49. [57]

    Learning 3d representations from 2d pre-trained models via image-to-point masked autoencoders

    Renrui Zhang, Liuhui Wang, Yu Qiao, Peng Gao, and Hong- sheng Li. Learning 3d representations from 2d pre-trained models via image-to-point masked autoencoders. InPro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 21769–21780, Van...

    2023

  50. [58]

    Parameter- efficient and memory-efficient tuning for vision transformer: A disentangled approach.arXiv preprint arXiv:2407.06964,

    Taolin Zhang, Jiawang Bai, Zhihe Lu, Dongze Lian, Gen- ping Wang, Xinchao Wang, and Shu-Tao Xia. Parameter- efficient and memory-efficient tuning for vision transformer: A disentangled approach.arXiv preprint arXiv:2407.06964,

    arXiv

  51. [59]

    Point cloud pre-training with diffusion models

    Xiao Zheng, Xiaoshui Huang, Guofeng Mei, Yuenan Hou, Zhaoyang Lyu, Bo Dai, Wanli Ouyang, and Yongshun Gong. Point cloud pre-training with diffusion models. InProceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 22935–22945, 2024. 6

    2024

  52. [60]

    Conditional prompt learning for vision-language mod- els

    Kaiyang Zhou, Jingkang Yang, Chen Change Loy, and Ziwei Liu. Conditional prompt learning for vision-language mod- els. InProceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition, pages 16816–16825,

  53. [61]

    Learning to prompt for vision-language models.In- ternational Journal of Computer Vision, 130(9):2337–2348,

    Kaiyang Zhou, Jingkang Yang, Chen Change Loy, and Ziwei Liu. Learning to prompt for vision-language models.In- ternational Journal of Computer Vision, 130(9):2337–2348,

  54. [62]

    Dynamic adapter meets prompt tuning: Parameter-efficient transfer learning for point cloud analysis

    Xin Zhou, Dingkang Liang, Wei Xu, Xingkui Zhu, Yihan Xu, Zhikang Zou, and Xiang Bai. Dynamic adapter meets prompt tuning: Parameter-efficient transfer learning for point cloud analysis. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages ...

    2024

  55. [63]

    Vision mamba: Efficient visual representation learning with bidirectional state space model.arXiv preprint arXiv:2401.09417, 2024

    Lianghui Zhu, Bencheng Liao, Qian Zhang, Xinlong Wang, Wenyu Liu, and Xinggang Wang. Vision mamba: Efficient visual representation learning with bidirectional state space model.arXiv preprint arXiv:2401.09417, 2024. 2

    2024 arXiv

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.