Pith. sign in

REVIEW 3 major objections 5 minor 68 references

MambaPro: Multi-Modal Object Re-Identification with Mamba Aggregation and Synergistic Prompt

T0 review · 3 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read MambaPro freezes CLIP, adds parallel adapters, synergistic prompts, and Mamba fusion, and reports state-of-the-art mAP on three multi-modal ReID benchmarks.

desk verdict Solid PEFT+CLIP+Mamba ReID paper with clean ablations; SOTA claim is confounded by CLIP pretraining and needs reconditioning before acceptance. read the letter →

arxiv 2412.10707 v1 pith:K56TKK65 submitted 2024-12-14 cs.CV cs.MM

Yuhao Wang , Xuehu Liu , Tianyu Yan , Yang Liu , Aihua Zheng , Pingping Zhang , Huchuan Lu This is my paper · ORCID
classification cs.CVcs.MM
keywords multi-modalobjectre-identificationparameter-efficientfine-tuningCLIPadaptationMambastatespacemodelprompttuningadapterRGB-NIR-TIRfusionbenchmarks
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

Multi-modal object re-identification (ReID) tries to find the same person or vehicle across cameras when images come from RGB, near-infrared, and thermal sensors. MambaPro's central claim is that a frozen CLIP image encoder, left untouched, can beat full fine-tuning on this task once three small trainable modules are attached: a Parallel Feed-Forward Adapter (PFA), a Synergistic Residual Prompt (SRP) that exchanges prompt tokens between modalities, and a Mamba Aggregation (MA) that models within- and across-modality token sequences with linear-complexity state space blocks. On the three benchmarks the paper reports the best mean average precision (mAP) so far: 78.9% on RGBNT201, 83.9% on RGBNT100, and 47.0% on MSVR310, with a trainable-parameter count well below full fine-tuning. The reason a sympathetic reader should care is that this points toward cheap reuse of general-purpose vision-language models for multi-modal retrieval, and toward state space models as a substitute for quadratic self-attention when fusing long multi-modal sequences.

What carries the argument

The load-bearing mechanism is the interaction of the frozen CLIP image encoder with three trainable side modules. The Parallel Feed-Forward Adapter (PFA) is a linear-GELU-linear branch connected in parallel with each frozen feed-forward block; it preserves the original feature flow while transferring CLIP knowledge to the ReID task. The Synergistic Residual Prompt (SRP) concatenates prompt tokens into each layer's input, transforms each modality's prompts into the other modalities via a transfer block, and adds the average of previous prompts to the next layer's prompts, so multi-modal information is exchanged and carried forward. The Mamba Aggregation (MA) uses selective state space models (SSMs), the mechanism behind Mamba that processes sequences in linear rather than quadratic time, to model intra-modality token interactions and then inter-modality interactions over the concatenated three-modality token sequence.

What would settle it

Train the strongest prior architecture (for example TOP-ReID) with the same frozen CLIP image encoder and training recipe but without PFA/SRP/MA, and measure RGBNT201 mAP; if it reaches the 78.9% that MambaPro reports, the claimed module-level gains are not responsible for the headline result. The paper's own supplementary runs this control for TOP-ReID under full fine-tuning and records 73.3% mAP, so a frozen-backbone control is the missing experiment.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that a strong multi-modal ReID model does not need to fine-tune a large pre-trained transformer; it needs to preserve the pre-trained representations while adding modality-aware information exchange around them. The PFA sits parallel to each feed-forward block so original features flow through unchanged, while an ascending-then-descending linear adapter injects ReID-specific refinements. The SRP makes each modality's prompt tokens learnable, transfers them into the other modalities through a linear-GELU-linear transfer block, and carries fused prompts forward with residual connections so no layer's prompt information is discarded. The MA then processes patch tokens of each modality separately through a selective state space block, concatenates tokens from all modalities into one long sequence, and processes that sequence through another state space block before splitting it back. With these parts, MambaPro reports 78.9% mAP and 83.4% Rank-1 on RGBNT201, 83.9% mAP on RGBNT100, and 47.0% mAP on MSVR310.

Load-bearing premise

The central comparison assumes MambaPro's gains come from its modules, but the prior methods in the main tables were not trained with CLIP initialization; if CLIP's pre-trained features alone explain most of the gap, the architectural advantage would shrink.

Editorial extensions

If this is right

  • Full fine-tuning of a large pre-trained CLIP encoder is not necessary for multi-modal ReID: the frozen-backbone model (78.9% mAP) matches the fully fine-tuned variant (78.7%) while training far fewer parameters.
  • Mamba blocks can replace transformer attention in the fusion stage: the Mamba aggregator reaches 78.9% mAP versus 76.5% for a transformer aggregator at roughly comparable FLOPs.
  • Cross-modal prompt transfer works as a fusion mechanism: transferring prompt tokens between modalities beats independent per-modality prompts (68.9 vs 64.7 mAP), and adding the residual prompt fusion pushes this further (71.3 mAP).
  • Where the adapter is placed matters: a parallel branch to the feed-forward network gives 73.2 mAP, whereas inserting it after the feed-forward block drops to 46.9 mAP.
  • CLIP pre-training itself matters in this setup: the frozen CLIP backbone reaches 78.9 mAP while a frozen ImageNet-pretrained ViT reaches 67.5 mAP.

Reading between the lines

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

  • Editorial inference: because all main-table comparisons train prior methods without CLIP initialization, the size of the reported margin is partly a statement about CLIP pretraining, not only about the new modules; the supplement controls this for TOP-ReID only, where the CLIP-based comparison is 73.3 vs 78.7 mAP.
  • Editorial inference: the linear-complexity claim applies to the Mamba aggregation stage; the frozen CLIP backbone itself still runs quadratic self-attention over its patch tokens, so the whole pipeline is not linear end to end.
  • Editorial inference: the synergistic prompt design could be lifted into other multi-modal retrieval settings, such as text-based person search or audio-visual matching, wherever each modality has its own set of prompts.
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 / 5 minor

Summary. The paper introduces MambaPro, a parameter-efficient fine-tuning framework for multi-modal object re-identification. It uses a frozen CLIP image encoder with three proposed modules: a Parallel Feed-Forward Adapter (PFA) for knowledge transfer, a Synergistic Residual Prompt (SRP) for modality interaction, and a Mamba Aggregation (MA) block for intra- and inter-modality sequence modeling with claimed linear complexity. Experiments on RGBNT201, RGBNT100, and MSVR310 report state-of-the-art mAP, and ablations show each module contributes over a frozen CLIP baseline. The main empirical claim is that MambaPro surpasses prior multi-modal ReID methods, e.g., by 6.6 mAP over TOP-ReID on RGBNT201, while using fewer trainable parameters than full fine-tuning.

Significance. If validated, the contribution is significant for the multi-modal ReID community: it demonstrates that a frozen CLIP backbone can be adapted to multi-modal object ReID with lightweight modules, and the internal ablation design is careful, with each proposed component tested over the same frozen-backbone baseline. The released code is a practical asset. However, the headline state-of-the-art claim is currently confounded by the CLIP pretraining of MambaPro versus the non-CLIP baselines in the main tables; the internal ablations are convincing, but the externally valid comparison needs additional controlled baselines.

major comments (3)
  1. [Performance Comparison (Tables 1-2) and Supplementary Tables 9-10] The state-of-the-art claim is not controlled for pretraining. Supplementary Table 9 shows that replacing the frozen ImageNet ViT backbone with frozen CLIP, while keeping all proposed modules fixed, raises RGBNT201 mAP from 67.5 to 78.9 (11.4 points), which is larger than the 6.6-point margin over TOP-ReID reported in Table 1. Since none of the baselines in Tables 1-2 uses CLIP initialization, the reported superiority cannot be attributed to the proposed modules alone. The only CLIP-based baseline, TOP-ReID+CLIP in Table 10, is compared against fully fine-tuned MambaPro (160.34M parameters, Model G in Table 3) rather than the proposed efficient MambaPro (74.20M, Model F), and this comparison is absent from the main tables. Please add CLIP-initialized versions of at least TOP-ReID, EDITOR, and RSCNet to Tables 1-2, or to a clearly labeled separate table, using comparable training recipes, and state explicitly which comparisons are CLIP-controlled.
  2. [Ablation Study, Tables 3-5] The parameter counts for the same configuration are inconsistent across tables. CLIP(L)+PFA is reported as 28.34M trainable parameters in Table 3 but 2.36M in Table 5, and CLIP(L)+SRP(Fusion) is 28.48M in Table 3 but 2.37M in Table 4. This discrepancy directly undermines the parameter-efficiency claims, including the statement in Section 'Effect of Key Components' that Model C uses '32.9% of the trainable parameters compared with Model B'. Please clarify what each column counts (module-only vs total trainable parameters, including any unfrozen LayerNorm, position embeddings, or classification head) and make the numbers consistent across all tables.
  3. [Mamba Aggregation and Complexity Claims (Tables 6, 11; Fig. 1)] The abstract claims 'lower FLOPs' and 'linear complexity' for the full framework, but the module-level numbers in Table 6 show the full MA block at 1.42G FLOPs and 9.47M parameters versus 1.36G and 4.73M for the single Transformer baseline used in the same table. The fair matched-structure comparison is deferred to supplementary Table 11 (MA(Trans) 1.81G/14.19M vs MA(Mamba) 1.42G/9.47M). Please bring the matched-structure comparison into the main text and report total model FLOPs for the full MambaPro compared with full-fine-tuning baselines, so that 'lower complexity' is substantiated by the actual empirical setup rather than by the asymptotic scaling argument in Fig. 1.
minor comments (5)
  1. [Implementation Details] There is a typo: 'RGBNT201/MVSR310' should be 'RGBNT201/MSVR310'.
  2. [Table 2] On RGBNT100, MambaPro's Rank-1 (94.7) is lower than several baselines (LRMM 96.7, CCNet 96.3, UniCat 96.2, TOP-ReID 96.4, EDITOR 96.4, RSCNet 96.6). The text only claims mAP superiority, but the abstract's 'surpassing TOP-ReID by 6.6% and 6.8%' refers to RGBNT201; please specify that the state-of-the-art claim is on mAP for RGBNT100 (and on both metrics for RGBNT201 and MSVR310), or provide additional evidence for a broader claim.
  3. [Tables 3, 7, 10] The name 'MambaPro' is used for two different models: the default PEFT variant with 74.20M parameters (Model F, Table 3) and the fully fine-tuned variant with 160.34M parameters (Model G, Table 10). Please disambiguate these variants in the text and tables to avoid confusion about which model is being compared.
  4. [Equations (8)-(12)] The notation for transferred prompts, e.g., \p^{r→n}_1], is used in the equations but not explicitly defined in the method text; define all prompt tokens and the transfer block T before using them in the formulas.
  5. [Figure 1] Figure 1(b) labels the comparison as 'SSM vs SA' but the actual empirical FLOPs comparison in Table 6 uses a Transformer aggregation block, not self-attention in the same dual intra/inter structure; consider aligning the figure's claim with the matched-structure experiment in Table 11.

Circularity Check

0 steps flagged · score 0.0 of 10

Empirical ReID paper with no definitional or self-citation circularity; the CLIP-initialization confound is an external-validity issue, not a circular step.

full rationale

I reviewed the method equations (PFA in Eqs. 1-3, SRP in Eqs. 4-12, MA in Eqs. 13-19), the objective (Eqs. 20-21), and all main and supplementary experiments. No proposed module is defined in terms of the reported metric, and no fitted parameter is renamed as a prediction. The central claims are supported by controlled ablations in Tables 3-6 and 12, where components are added or removed under the same frozen CLIP backbone, and by the backbone comparison in Supp. Table 9. The main limitation is that the headline state-of-the-art comparison in Tables 1-2 uses ViT- or CNN-based baselines without CLIP initialization, so part of the margin may come from the pretrained backbone rather than the proposed modules; however, this is an experimental-design/external-validity concern, not a circularity, because CLIP is an external pretrained model and the supplement provides a CLIP-initialized TOP-ReID comparison (Table 10) where MambaPro still leads. Some hyperparameters are selected after observing benchmark performance, which is a reproducibility risk, but the paper does not present those selections as independent predictions, so it does not fit the fitted-input-called-prediction pattern. Self-citations to prior works such as TOP-ReID and Magic tokens provide baseline numbers and context, but no load-bearing argument reduces to those citations. I find no circular step that quotes and exhibits a reduction of the claimed result to its inputs.

Assumptions & free parameters 7 free parameters · 3 assumptions · 0 invented entities

No new physical entities are introduced. The paper's reliance is on pre-trained CLIP and on tuned hyperparameters.

free parameters (7)
  • dstate = 16
    Hidden state dimension in SSM; tuned on RGBNT201, performance peaks at 16 (Fig. 8a).
  • dtrank = not stated (chosen >32)
    Rank for generating discrete time steps in SSM; performance rises when dtrank > 32 (Fig. 8b). Exact value not reported.
  • Npr = 4
    Number of prompt tokens per modality, tuned to 4 in Fig. 9a.
  • number of MA blocks = 2
    Two stacked MA blocks chosen as best trade-off (Fig. 9d); four blocks yield 79.5% mAP but higher cost.
  • lambda1 = 0.25
    Weight for cross-entropy loss in Eq. (20).
  • lambda2 = 1.0
    Weight for triplet loss in Eq. (20).
  • learning rates = 3.5e-4 (PEFT), 5e-6 (full fine-tune)
    Hyperparameters; no sensitivity analysis reported.
assumptions (3)
  • domain assumption CLIP's frozen image encoder can be adapted to multi-modal ReID via parallel adapters and prompts without catastrophic forgetting.
    This is the premise for the PFA and SRP designs, validated empirically but not theoretically.
  • domain assumption Mamba/SSM with the chosen scanning order can model cross-modal interactions as effectively as self-attention at the sequence lengths used.
    Assumed from Mamba's long-sequence capability; only empirically tested on three datasets.
  • domain assumption The three benchmarks' standard splits and protocols are followed, and reported prior-method numbers are accurate.
    The paper relies on previously published numbers for comparison; if those are inconsistent, the SOTA claim is unsupported.

how reviews work

0 comments
Cite this review

Pith. "Pith review of MambaPro: Multi-Modal Object Re-Identification with Mamba Aggregation and Synergistic Prompt." pith.science (2026). https://pith.science/paper/K56TKK65

@misc{pith2026241210707,
  author       = {Pith},
  title        = {Pith review of: MambaPro: Multi-Modal Object Re-Identification with Mamba Aggregation and Synergistic Prompt},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/K56TKK65}},
  note         = {Machine review of arXiv:2412.10707}
}
read the original abstract

Multi-modal object Re-IDentification (ReID) aims to retrieve specific objects by utilizing complementary image information from different modalities. Recently, large-scale pre-trained models like CLIP have demonstrated impressive performance in traditional single-modal object ReID tasks. However, they remain unexplored for multi-modal object ReID. Furthermore, current multi-modal aggregation methods have obvious limitations in dealing with long sequences from different modalities. To address above issues, we introduce a novel framework called MambaPro for multi-modal object ReID. To be specific, we first employ a Parallel Feed-Forward Adapter (PFA) for adapting CLIP to multi-modal object ReID. Then, we propose the Synergistic Residual Prompt (SRP) to guide the joint learning of multi-modal features. Finally, leveraging Mamba's superior scalability for long sequences, we introduce Mamba Aggregation (MA) to efficiently model interactions between different modalities. As a result, MambaPro could extract more robust features with lower complexity. Extensive experiments on three multi-modal object ReID benchmarks (i.e., RGBNT201, RGBNT100 and MSVR310) validate the effectiveness of our proposed methods. The source code is available at https://github.com/924973292/MambaPro.

Figures

Figures reproduced from arXiv: 2412.10707 by the authors.

Figure 1
Figure 1. (a) Comparison between previous methods and [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. The overall framework of MambaPro. The Parallel Feed-Forward Adapter (PFA) is first introduced to transfer pre [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Details of our proposed Mamba Aggregation. [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (8 more)
Figure 4
Figure 4. Figure 4: Details of different prompt mechanisms. Otherwise, we use fma for retrieval. Effect of Key Components. Tab. 3 shows the performance comparison with different components. Comparing Model A with Model B, we observe an impressive mAP improve￾ment by fully fine-tuning CLIP…
Figure 5
Figure 5. Figure 5: Feature distributions with t-SNE (Van der Maaten and Hinton 2008). Different colors represent different IDs. [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Alignment visualization of cosine similarity distri [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7: Detailed workflow of our SRP. Workflow of the SRP As shown in [PITH_FULL_IMAGE:figures/full_fig_p010_7.png]
Figure 8
Figure 8. Figure 8: Performance comparison with different dstate and dtrank in MA. (a) (b) (c) (d) PFA PFA [PITH_FULL_IMAGE:figures/full_fig_p012_8.png]
Figure 9
Figure 9. Figure 9: Comparison of different depths or prompt lengths in proposed modules. [PITH_FULL_IMAGE:figures/full_fig_p012_9.png]
Figure 10
Figure 10. Figure 10: Grad-CAM visualizations of multi-modal person ReID. [PITH_FULL_IMAGE:figures/full_fig_p013_10.png]
Figure 11
Figure 11. Figure 11: Grad-CAM visualizations of multi-modal vehicle ReID. [PITH_FULL_IMAGE:figures/full_fig_p013_11.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

68 extracted references · 39 canonical work pages

  1. [1]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.a...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

  3. [3]

    L.; Kiros, J

    Ba, J. L.; Kiros, J. R.; and Hinton, G. E. 2016. Layer normalization. arXiv preprint arXiv:1607.06450

    arXiv 2016

  4. [4]

    M.; and Xiang, T

    Chang, X.; Hospedales, T. M.; and Xiang, T. 2018. Multi-level factorisation net for person re-identification. In CVPR, 2109--2118

    work page 2018

  5. [5]

    Chollet, F. 2017. Xception: Deep learning with depthwise separable convolutions. In CVPR, 1251--1258

    work page 2017

  6. [6]

    Crawford, J.; Yin, H.; McDermott, L.; and Cummings, D. 2023. UniCat: Crafting a Stronger Fusion Baseline for Multimodal Re-Identification. arXiv preprint arXiv:2310.18812

    arXiv 2023

  7. [7]

    Diao, H.; Wan, B.; Jia, X.; Zhuge, Y.; Zhang, Y.; Lu, H.; and Chen, L. 2024 a . Sherl: Synthesizing high accuracy and efficient memory for resource-limited transfer learning. In ECCV, 75--95

    work page 2024

  8. [8]

    Diao, H.; Wan, B.; Zhang, Y.; Jia, X.; Lu, H.; and Chen, L. 2024 b . Unipt: Universal parallel tuning for transfer learning with efficient parameter and memory. In CVPR, 28729--28740

    work page 2024

Show all 68 references
  1. [9]

    Dosovitskiy, A.; Beyer, L.; Kolesnikov, A.; Weissenborn, D.; Zhai, X.; Unterthiner, T.; Dehghani, M.; Minderer, M.; Heigold, G.; Gelly, S.; et al. 2020. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv:2010.11929

    2020 arXiv

  2. [10]

    Elfwing, S.; Uchibe, E.; and Doya, K. 2018. Sigmoid-weighted linear units for neural network function approximation in reinforcement learning. Neural networks, 107: 3--11

    2018

  3. [11]

    French, R. M. 1999. Catastrophic forgetting in connectionist networks. Trends in cognitive sciences, 3(4): 128--135

    1999

  4. [12]

    Gu, A.; and Dao, T. 2023. Mamba: Linear-time sequence modeling with selective state spaces. arXiv preprint arXiv:2312.00752

    2023 arXiv

  5. [13]

    Gu, A.; Goel, K.; and R \'e , C. 2021. Efficiently modeling long sequences with structured state spaces. arXiv preprint arXiv:2111.00396

    2021 arXiv

  6. [14]

    Gu, A.; Johnson, I.; Goel, K.; Saab, K.; Dao, T.; Rudra, A.; and R \'e , C. 2021. Combining recurrent, convolutional, and continuous-time models with linear state space layers. NeurIPS, 34: 572--585

    2021

  7. [15]

    Guo, J.; Zhang, X.; Liu, Z.; and Wang, Y. 2022. Generative and attentive fusion for multi-spectral vehicle re-identification. In ICSP, 1565--1572

    2022

  8. [16]

    He, Q.; Lu, Z.; Wang, Z.; and Hu, H. 2023. Graph-Based Progressive Fusion Network for Multi-Modality Vehicle Re-Identification. TITS, 1--17

    2023

  9. [17]

    He, S.; Luo, H.; Wang, P.; Wang, F.; Li, H.; and Jiang, W. 2021. Transreid: Transformer-based object re-identification. In ICCV, 15013--15022

    2021

  10. [18]

    Hendrycks, D.; and Gimpel, K. 2016. Gaussian error linear units (gelus). arXiv preprint arXiv:1606.08415

    2016 arXiv

  11. [19]

    Hermans, A.; Beyer, L.; and Leibe, B. 2017. In defense of the triplet loss for person re-identification. arXiv preprint arXiv:1703.07737

    2017 arXiv

  12. [20]

    Houlsby, N.; Giurgiu, A.; Jastrzebski, S.; Morrone, B.; De Laroussilhe, Q.; Gesmundo, A.; Attariyan, M.; and Gelly, S. 2019. Parameter-efficient transfer learning for NLP. In ICML, 2790--2799

    2019

  13. [21]

    J.; Shen, Y.; Wallis, P.; Allen-Zhu, Z.; Li, Y.; Wang, S.; Wang, L.; and Chen, W

    Hu, E. J.; Shen, Y.; Wallis, P.; Allen-Zhu, Z.; Li, Y.; Wang, S.; Wang, L.; and Chen, W. 2021. Lora: Low-rank adaptation of large language models. arXiv preprint arXiv:2106.09685

    2021 arXiv

  14. [22]

    Ioffe, S.; and Szegedy, C. 2015. Batch normalization: Accelerating deep network training by reducing internal covariate shift. In ICML, 448--456

    2015

  15. [23]

    M.; and Bertasius, G

    Islam, M. M.; and Bertasius, G. 2022. Long movie clip classification with state-space video models. In ECCV, 87--104

    2022

  16. [24]

    M.; Hasan, M.; Athrey, K

    Islam, M. M.; Hasan, M.; Athrey, K. S.; Braskich, T.; and Bertasius, G. 2023. Efficient Movie Scene Detection using State-Space Transformers. In CVPR, 18749--18758

    2023

  17. [25]

    U.; Rasheed, H.; Maaz, M.; Khan, S.; and Khan, F

    Khattak, M. U.; Rasheed, H.; Maaz, M.; Khan, S.; and Khan, F. S. 2023. Maple: Multi-modal prompt learning. In CVPR, 19113--19122

    2023

  18. [26]

    Lester, B.; Al-Rfou, R.; and Constant, N. 2021. The power of scale for parameter-efficient prompt tuning. arXiv preprint arXiv:2104.08691

    2021 arXiv

  19. [27]

    Li, H.; Li, C.; Zhu, X.; Zheng, A.; and Luo, B. 2020. Multi-spectral vehicle re-identification: A challenge. In AAAI, volume 34, 11345--11353

    2020

  20. [28]

    Li, S.; Sun, L.; and Li, Q. 2023. CLIP-ReID: exploiting vision-language model for image re-identification without concrete text labels. In AAAI, volume 37, 1405--1413

    2023

  21. [29]

    Li, W.; Zhu, X.; and Gong, S. 2018. Harmonious attention network for person re-identification. In CVPR, 2285--2294

    2018

  22. [30]

    Li, Y.; Liu, Z.; Yang, W.; Wang, Y.; Liao, Q.; et al. 2023. CLIP-based Synergistic Knowledge Transfer for Text-based Person Retrieval. arXiv preprint arXiv:2309.09496

    2023 arXiv

  23. [31]

    W.; Zhang, Y.; Kwon, Y.; Yeung, S.; and Zou, J

    Liang, V. W.; Zhang, Y.; Kwon, Y.; Yeung, S.; and Zou, J. Y. 2022. Mind the gap: Understanding the modality gap in multi-modal contrastive representation learning. NeurIPS, 35: 17612--17625

    2022

  24. [32]

    Liu, X.; Yu, C.; Zhang, P.; and Lu, H. 2023. Deeply coupled convolution--transformer with spatial--temporal complementary learning for video-based person re-identification. TNNLS

    2023

  25. [33]

    Liu, X.; Zhang, P.; Yu, C.; Lu, H.; and Yang, X. 2021. Watching you: Global-guided reciprocal learning for video-based person re-identification. In CVPR, 13334--13343

    2021

  26. [34]

    Liu, X.; Zhang, P.; Yu, C.; Qian, X.; Yang, X.; and Lu, H. 2024 a . A video is worth three views: Trigeminal transformers for video-based person re-identification. TITS, 25

    2024

  27. [35]

    Liu, Y.; Tian, Y.; Zhao, Y.; Yu, H.; Xie, L.; Wang, Y.; Ye, Q.; and Liu, Y. 2024 b . Vmamba: Visual state space model. arXiv preprint arXiv:2401.10166

    2024 arXiv

  28. [36]

    Luo, H.; Gu, Y.; Liao, X.; Lai, S.; and Jiang, W. 2019. Bag of tricks and a strong baseline for deep person re-identification. In CVPRW, 1487--1495

    2019

  29. [37]

    Pan, W.; Huang, L.; Liang, J.; Hong, L.; and Zhu, J. 2023. Progressively Hybrid Transformer for Multi-Modal Vehicle Re-Identification. Sensors, 23(9): 4206

    2023

  30. [38]

    Qian, X.; Fu, Y.; Jiang, Y.-G.; Xiang, T.; and Xue, X. 2017. Multi-scale deep learning architectures for person re-identification. In ICCV, 5399--5408

    2017

  31. [39]

    W.; Hallacy, C.; Ramesh, A.; Goh, G.; Agarwal, S.; Sastry, G.; Askell, A.; Mishkin, P.; Clark, J.; et al

    Radford, A.; Kim, J. W.; Hallacy, C.; Ramesh, A.; Goh, G.; Agarwal, S.; Sastry, G.; Askell, A.; Mishkin, P.; Clark, J.; et al. 2021. Learning transferable visual models from natural language supervision. In ICML, 8748--8763

    2021

  32. [40]

    Rao, Y.; Chen, G.; Lu, J.; and Zhou, J. 2021. Counterfactual attention learning for fine-grained visual categorization and re-identification. In ICCV, 1025--1034

    2021

  33. [41]

    Shi, J.; Yin, X.; Chen, Y.; Zhang, Y.; Zhang, Z.; Xie, Y.; and Qu, Y. 2024 a . Multi-Memory Matching for Unsupervised Visible-Infrared Person Re-Identification. arXiv preprint arXiv:2401.06825

    2024 arXiv

  34. [42]

    Shi, J.; Yin, X.; Zhang, Y.; Xie, Y.; Qu, Y.; et al. 2024 b . Learning commonality, divergence and variety for unsupervised visible-infrared person re-identification. In NeurIPS

    2024

  35. [43]

    T.; Warrington, A.; and Linderman, S

    Smith, J. T.; Warrington, A.; and Linderman, S. W. 2022. Simplified state space layers for sequence modeling. arXiv preprint arXiv:2208.04933

    2022 arXiv

  36. [44]

    Sun, Y.; Zheng, L.; Yang, Y.; Tian, Q.; and Wang, S. 2018. Beyond part models: Person retrieval with refined part pooling (and a strong convolutional baseline). In ECCV, 480--496

    2018

  37. [45]

    Szegedy, C.; Vanhoucke, V.; Ioffe, S.; Shlens, J.; and Wojna, Z. 2016. Rethinking the inception architecture for computer vision. In CVPR, 2818--2826

    2016

  38. [46]

    Van der Maaten, L.; and Hinton, G. 2008. Visualizing data using t-SNE. JMLR, 9(11)

    2008

  39. [47]

    N.; Kaiser, .; and Polosukhin, I

    Vaswani, A.; Shazeer, N.; Parmar, N.; Uszkoreit, J.; Jones, L.; Gomez, A. N.; Kaiser, .; and Polosukhin, I. 2017. Attention is all you need. NeurIPS, 30

    2017

  40. [48]

    Wan, Z.; Zhang, P.; Wang, Y.; Yong, S.; Stepputtis, S.; Sycara, K.; and Xie, Y. 2024. Sigma: Siamese mamba network for multi-modal semantic segmentation. arXiv preprint arXiv:2404.04256

    2024 arXiv

  41. [49]

    Wang, Y.; Liu, X.; Zhang, P.; Lu, H.; Tu, Z.; and Lu, H. 2023. TOP-ReID: Multi-spectral Object Re-Identification with Token Permutation. arXiv preprint arXiv:2312.09612

    2023 arXiv

  42. [50]

    Wang, Y.; Zhang, P.; Wang, D.; and Lu, H. 2024 a . Other tokens matter: Exploring global and local features of Vision Transformers for Object Re-Identification. CVIU, 244: 104030

    2024

  43. [51]

    Wang, Z.; Huang, H.; Zheng, A.; and He, R. 2024 b . Heterogeneous Test-Time Training for Multi-Modal Person Re-identification. In AAAI, volume 38, 5850--5858

    2024

  44. [52]

    Wang, Z.; Li, C.; Zheng, A.; He, R.; and Tang, J. 2022. Interact, embed, and enlarge: Boosting modality-specific representations for multi-modal person re-identification. In AAAI, volume 36, 2633--2641

    2022

  45. [53]

    Wu, D.; Liu, Z.; Chen, Z.; Gan, S.; Tan, K.; Wan, Q.; and Wang, Y. 2025. LRMM: Low rank multi-scale multi-modal fusion for person re-identification based on RGB-NI-TI. ESWA, 263: 125716

    2025

  46. [54]

    Xu, Z.; Tang, F.; Chen, Z.; Zhou, Z.; Wu, W.; Yang, Y.; Liang, Y.; Jiang, J.; Cai, X.; and Su, J. 2024. Polyp-Mamba: Polyp Segmentation with Visual Mamba. In MICCAI

    2024

  47. [55]

    Yang, B.; Chen, J.; and Ye, M. 2023. Towards Grand Unified Representation Learning for Unsupervised Visible-Infrared Person Re-Identification. In ICCV, 11069--11079

    2023

  48. [56]

    Yang, B.; Chen, J.; and Ye, M. 2024. Shallow-Deep Collaborative Learning for Unsupervised Visible-Infrared Person Re-Identification. In CVPR, 16870--16879

    2024

  49. [57]

    Yang, L.; Kang, B.; Huang, Z.; Xu, X.; Feng, J.; and Zhao, H. 2024. Depth anything: Unleashing the power of large-scale unlabeled data. arXiv preprint arXiv:2401.10891

    2024 arXiv

  50. [58]

    Yin, H.; Li, J.; Schiller, E.; McDermott, L.; and Cummings, D. 2023. GraFT: Gradual Fusion Transformer for Multimodal Re-Identification. arXiv preprint arXiv:2310.16856

    2023 arXiv

  51. [59]

    Yu, C.; Liu, X.; Wang, Y.; Zhang, P.; and Lu, H. 2024 a . TF-CLIP: Learning text-free CLIP for video-based person re-identification. In AAAI, volume 38, 6764--6772

    2024

  52. [60]

    Yu, Z.; Huang, Z.; Hou, M.; Pei, J.; Yan, Y.; Liu, Y.; and Sun, D. 2024 b . Representation Selective Coupling via Token Sparsification for Multi-Spectral Object Re-Identification. TCSVT

    2024

  53. [61]

    Zhang, G.; Zhang, P.; Qi, J.; and Lu, H. 2021. Hat: Hierarchical aggregation transformers for person re-identification. In ACM MM, 516--525

    2021

  54. [62]

    Zhang, P.; Wang, Y.; Liu, Y.; Tu, Z.; and Lu, H. 2024. Magic tokens: Select diverse tokens for multi-modal object re-identification. In CVPR, 17117--17126

    2024

  55. [63]

    Zheng, A.; He, Z.; Wang, Z.; Li, C.; and Tang, J. 2023. Dynamic Enhancement Network for Partial Multi-modality Person Re-identification. arXiv preprint arXiv:2305.15762

    2023 arXiv

  56. [64]

    Zheng, A.; Wang, Z.; Chen, Z.; Li, C.; and Tang, J. 2021. Robust multi-modality person re-identification. In AAAI, volume 35, 3529--3537

    2021

  57. [65]

    Zheng, A.; Zhu, X.; Ma, Z.; Li, C.; Tang, J.; and Ma, J. 2022. Multi-spectral vehicle re-identification with cross-directional consistency network and a high-quality benchmark. arXiv preprint arXiv:2208.00632

    2022 arXiv

  58. [66]

    Zhong, Z.; Zheng, L.; Kang, G.; Li, S.; and Yang, Y. 2020. Random erasing data augmentation. In AAAI, volume 34, 13001--13008

    2020

  59. [67]

    Zhou, K.; Yang, Y.; Cavallaro, A.; and Xiang, T. 2019. Omni-scale feature learning for person re-identification. In ICCV, 3702--3712

    2019

  60. [68]

    Zhu, L.; Liao, B.; Zhang, Q.; Wang, X.; Liu, W.; and Wang, X. 2024. Vision mamba: Efficient visual representation learning with bidirectional state space model. arXiv preprint arXiv:2401.09417

    2024 arXiv

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.