Pith. sign in

REVIEW 4 major objections 1 minor 2 cited by

Onsager Principle-Based Domain Embedding for Thermodynamically Consistent Cahn-Hilliard Model in Arbitrary Domain

T0 review · 4 major / 1 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read This paper claims that the Cahn-Hilliard model on an arbitrary domain with two prescribed boundary conditions can be embedded into a larger, regular domain with homogeneous Neumann boundary conditions via a modified conservation law and the

desk verdict The abstract promises a useful domain-embedding result, but the supplied manuscript is an unrelated computer-vision paper, so there is no derivable content to referee. read the letter →

arxiv 2508.06830 v1 pith:MHU5NFP4 submitted 2025-08-09 math.NA cs.NA

Wenkai Yu , Qi Wang , Zhen Zhang , Tiezheng Qian This is my paper
classification math.NAcs.NA MSC 35K5565M0665M60
keywords Cahn-HilliardOnsagerprincipledomainembeddingthermodynamicconsistencyboundaryconditionsgradientflowphase-fieldstructure-preservingscheme
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

The paper tries to show that a Cahn-Hilliard phase-separation problem posed on an arbitrarily shaped domain, with two prescribed boundary conditions, can be solved instead as a modified Cahn-Hilliard-type problem on a larger, regular domain with simple no-flux boundary conditions. The key move is to change the conservation law so the flux crossing the original domain's boundary appears as a source term in the extended domain, and to derive the extended system's constitutive equation from the Onsager principle. The authors argue this works even when the rate at which free energy is pumped into the system is not known explicitly. If true, this gives a systematic way to handle gradient-flow problems in arbitrary geometries with thermodynamically consistent structure, and the asymptotic analysis indicates the original model and its boundary conditions are recovered in the embedding limit.

What carries the argument

The OPBDE method: a domain embedding procedure based on the Onsager variational principle, in which the conservation law is modified so the flux at the original domain's boundary becomes a source term in the extended domain. This object carries the argument by replacing a problem with two boundary conditions on an arbitrary boundary with a problem on a regular domain with homogeneous Neumann conditions, and the variational principle supplies the constitutive equation without explicit knowledge of the pumped free energy rate.

What would settle it

Examine the free-energy dissipation identity of the extended OPBDE Cahn-Hilliard model on a simple two-dimensional domain, for instance a circle embedded in a square. If the modified conservation law introduces a source term that makes the total free energy non-monotone, or if the asymptotic limit with vanishing regularization fails to reproduce both prescribed boundary conditions to the expected order, the central claim collapses. A concrete check is to compute the boundary-layer correction analytically for the linearized model and verify that both boundary conditions are recovered.

Watch

Extended reading notes

Core claim

The central claim is that the Onsager-principle-based domain embedding (OPBDE) method extends the original Cahn-Hilliard model in an arbitrary domain with two prescribed boundary conditions to a Cahn-Hilliard-type model in a larger regular domain with homogeneous Neumann boundary conditions, while preserving thermodynamic consistency. The modified conservation law inserts the original boundary flux as a source term; the Onsager principle still supplies the constitutive equation of the extended system even without prior knowledge of the rate of free energy pumped into the system. Asymptotic analysis demonstrates full recovery of the original model including its boundary conditions, and a stru

Load-bearing premise

The load-bearing premise is that the modified conservation law absorbs the original domain's boundary flux exactly as a source term, and that the Onsager variational structure survives in the extended system even though the rate of free energy pumped into the system is unknown; the asymptotic recovery of the two original boundary conditions also requires unstated convergence conditions on the regularization.

Editorial extensions

If this is right

  • Arbitrary-domain Cahn-Hilliard problems with two prescribed boundary conditions can be replaced by a thermodynamically consistent model on a larger regular domain with homogeneous Neumann boundary conditions.
  • The extended model can be solved with structure-preserving numerical schemes, making the approach usable as a practical computational tool for complex geometries.
  • Because the Onsager derivation does not require explicit knowledge of the pumped free-energy rate, the embedding framework applies even when the energy exchange between the original domain and its exterior is not fully characterized.
  • Asymptotic recovery of the original boundary conditions means the embedded solution can be viewed as a regularization of the original problem, with a well-defined limit.
  • The same OPBDE construction is positioned as a general strategy for gradient-flow problems in arbitrary domain geometries, going beyond the specific Cahn-Hilliard example.

Reading between the lines

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

  • If the embedding is as exact as claimed, the same modified-conservation-law construction should extend to other gradient flows with flux-type boundary conditions, such as Allen-Cahn or multi-component phase-field models.
  • The asymptotic recovery suggests a practical recipe: choose the embedding scale or regularization small relative to domain features; a natural test is to measure the convergence rate of the embedded solution to the original solution as the regularization vanishes.
  • A structure-preserving discretization on the larger regular domain could enable spectral or FFT-based solvers, making arbitrary-geometry phase-field simulations noticeably cheaper.
  • The claim that the Onsager derivation works without knowing the pumped-energy rate is testable in the linearized Cahn-Hilliard case, where the extended constitutive equation can be derived explicitly and compared with a direct variational calculation.
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

4 major / 1 minor

Summary. The manuscript as received consists of an abstract for an Onsager-principle-based domain embedding (OPBDE) method for the Cahn-Hilliard model, together with a full text that is an unrelated paper on low-rank expert merging for person re-identification (arXiv:2508.06831v2). The abstract claims a variational construction that embeds an arbitrary-domain Cahn-Hilliard problem with two prescribed boundary conditions into a larger regular domain with homogeneous Neumann conditions by absorbing the original boundary flux as a source term; claims that the Onsager variational structure survives without knowing the pumped free-energy rate; claims an asymptotic recovery of the original model including its boundary conditions; and claims numerical validation with a structure-preserving scheme. The supplied full text contains none of these elements: no Cahn-Hilliard equation, no Onsager principle, no modified conservation law, no asymptotic analysis, and no numerical scheme for the claimed model.

Significance. If the abstract's claims were fully substantiated, the work would be potentially significant for gradient flows and phase-field modeling on arbitrary domains, by offering a thermodynamically consistent embedding onto regular domains with homogeneous Neumann conditions. However, none of the necessary mathematical apparatus appears in the submitted text. There is no derivation, no theorem statement, no numerical experiment, and no code. The contribution is therefore not assessable on the evidence supplied. The manuscript cannot be considered a paper on the claimed topic in its current form.

major comments (4)
  1. [Full Text (arXiv:2508.06831v2)] The body of the submission is an unrelated computer vision manuscript on person re-identification. No equation in the body concerns the Cahn-Hilliard model; no Onsager principle, modified conservation law, source term, asymptotic expansion, or structure-preserving discretization appears. The abstract is the only place where the claimed mathematical result is stated. This is a complete evidentiary gap for the central claim and cannot be repaired by routine revision.
  2. [Abstract, 'modified conservation law'] The load-bearing modeling step—incorporating the flux at the original domain boundary as a source term in the extended domain's conservation law—is asserted without derivation. To verify thermodynamic consistency and the claimed Onsager structure, one needs an explicit definition of the source term, the extended free energy, the dissipation rate, and a proof that the variational principle remains valid despite the unknown pumped-energy rate. None of this is provided.
  3. [Abstract, 'asymptotic analysis'] The claim that the original Cahn-Hilliard model, including its boundary conditions, is 'fully recovered' requires a precise asymptotic statement: the limit parameter, the topology of convergence, assumptions on regularity, and a proof. The submitted text contains no such statement, so the recovery claim is unverifiable.
  4. [Abstract, 'structure-preserving numerical scheme'] The abstract promises a structure-preserving discretization and numerical results showing accuracy, effectiveness, and robustness. No scheme, no tables, no figures, and no convergence data for the claimed OPBDE Cahn-Hilliard model appear anywhere in the submission. The numerical validation cannot be checked.
minor comments (1)
  1. [Manuscript metadata] The title, abstract, and subject classification (math.NA) are inconsistent with the attached full text. If this is an administrative error, the correct body must be supplied; as it stands, all internal references (e.g., 'Sec. 3.3', Eq. (1)–(10)) refer to the unrelated re-identification paper.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity demonstrable: the supplied full text does not contain the claimed derivation, so the abstract's mathematical claims are unsupported but not shown to reduce to their inputs.

full rationale

The submitted manuscript's abstract claims an Onsager-principle-based domain embedding recovery of the Cahn-Hilliard model, but the supplied full text is arXiv:2508.06831v2 ('Low-Rank Expert Merging for Multi-Source Domain Adaptation in Person Re-Identification'), a computer-vision paper containing no Cahn-Hilliard equation, no modified conservation law, no Onsager variational principle, and no asymptotic analysis. This is a complete evidentiary gap that makes the claimed derivation unverifiable, not a circular reduction: there are no equations to compare, no fitted parameters renamed as predictions, and no self-citation chain to inspect. Under hard rule 1, a circularity finding requires quoting the paper and exhibiting the specific reduction; with no derivation present, no such reduction can be exhibited. The mismatch is therefore weighed as a correctness/verification failure rather than as evidence of circularity, and the circularity score is 0.

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

The ledger is abstract-only and therefore incomplete. The three listed axioms are the premises the abstract's central claim rests on: the source-term absorption of boundary flux, the validity of the Onsager structure without knowledge of the pumped free-energy rate, and the well-posedness of the asymptotic recovery limit. No free parameters are visible in the abstract: the construction does not mention fitted numbers, and the stated advantage is precisely that the pumped free-energy rate is not needed as input. No invented physical entities (new particles, forces, or dimensions) appear. Full audit requires the actual manuscript body, which was not supplied; the body that was supplied belongs to a different arXiv paper and is inadmissible for this audit.

assumptions (3)
  • ad hoc to paper Flux at the original domain's boundary can be incorporated into a modified conservation law as a source term on the extended domain.
    Abstract, modified conservation law sentence. This is the paper's central modeling construction; the entire extended model rests on it and it is not derived from outside the paper.
  • domain assumption The extended dissipative system obeys the Onsager variational principle even when the rate of free energy pumped into the system is unknown.
    Abstract, variational approach claim. Standard physics background for dissipative gradient-flow systems; the paper's derivation depends on it.
  • ad hoc to paper The embedding limit is well-behaved: the larger regular domain can be relaxed back to the original domain such that the source term reproduces the original boundary flux conditions.
    Abstract, asymptotic analysis claim. This is an assumption about convergence of the extended system to the original one, asserted rather than shown in the supplied text.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Onsager Principle-Based Domain Embedding for Thermodynamically Consistent Cahn-Hilliard Model in Arbitrary Domain." pith.science (2026). https://pith.science/paper/MHU5NFP4

@misc{pith2026250806830,
  author       = {Pith},
  title        = {Pith review of: Onsager Principle-Based Domain Embedding for Thermodynamically Consistent Cahn-Hilliard Model in Arbitrary Domain},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/MHU5NFP4}},
  note         = {Machine review of arXiv:2508.06830}
}
read the original abstract

The original Cahn-Hilliard model in an arbitrary domain with two prescribed boundary conditions is extended to a Cahn-Hilliard-type model in a larger, regular domain with homogeneous Neumann boundary conditions. The extension is based on the Onsager principle-based domain embedding (OPBDE) method, which has been developed as a systematic domain embedding framework to ensure thermodynamic consistency. By introducing a modified conservation law, the flux at the boundary of the original domain is incorporated into the conservation law as a source term. Our variational approach demonstrates that, even without a prior knowledge on the specific form of the rate of free energy pumped into the system, the Onsager principle remains an effective instrument in deriving the constitutive equation of the extended system. This approach clarifies the intrinsic structure of the extended model in the perspectives of free energy and its dissipation. Asymptotic analysis is carried out for the extended OPBDE Cahn-Hilliard model, demonstrating that the original Cahn-Hilliard model, including its boundary conditions, can be fully recovered. To validate our approach, a structure-preserving numerical scheme is developed to discretize the extended model. Numerical results show that the OPBDE Cahn-Hilliard model is accurate, effective, and robust, highlighting the capability of the OPBDE method in handling gradient flow problems in arbitrary domain geometries.

Discussion (0). Sign in to comment.

Forward citations

Cited by 2 Pith papers

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

  1. Onsager-variational formulation of diffuse-domain methods for computational modeling of microscale fluid-structure interactions

    physics.bio-ph 2026-05 unverdicted novelty 7.0 of 10

    Onsager-variational embedding of surface free-energy and dissipation functionals yields thermodynamically consistent diffuse-domain models for passive and active microscale fluid-structure interactions.

  2. Global Supply Chain Reallocation and Shift under Triple Crises: A U.S.-China Perspective

    econ.GN 2025-08 unverdicted novelty 5.0 of 10

    Even after three overlapping crises, U.S. imports shifted toward ASEAN 'China+1' partners whose supply chains remain tied to Chinese upstream inputs, so dependence on China-linked production did not decline.

Reference graph

Works this paper leans on

56 extracted references · 52 canonical work pages · cited by 2 Pith papers

  1. [1]

    Unsupervised multi-source domain adaptation for person re-identification

    Zechen Bai, Zhigang Wang, Jian Wang, Di Hu, and Er- rui Ding. Unsupervised multi-source domain adaptation for person re-identification. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 12914–12923, 2021. 1, 2, 5, 7, 8

    work page 2021

  2. [2]

    Buehler and Markus J

    Eric L. Buehler and Markus J. Buehler. X-lora: Mixture (a) Average gating coefficient (Target: DukeMTMC-reID) (b) Average Average gating coefficient (Target: Market-1501) (c) Average Average gating coefficient (Target: MSMT17) Figure 3. Average gating coefficient����� per target domain. of low-rank adapter experts, a flexible framework for large language ...

    work page

  3. [3]

    Domain-specific batch normalization for unsupervised domain adaptation

    Woong-Gi Chang, Tackgeun You, Seonguk Seo, Suha Kwak, and Bohyung Han. Domain-specific batch normalization for unsupervised domain adaptation. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 7354–7362, 2019. 2

    work page 2019

  4. [4]

    Idm: An intermediate domain mod- ule for domain adaptive person re-id

    Yongxing Dai, Jun Liu, Yifan Sun, Zekun Tong, Chi Zhang, and Ling-Yu Duan. Idm: An intermediate domain mod- ule for domain adaptive person re-id. InProceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pages 11864–11874, 2021. 1, 2, 5, 7

    work page 2021

  5. [5]

    Image-image domain adaptation with preserved self-similarity and domain-dissimilarity for person re-identification

    Weijian Deng, Liang Zheng, Qixiang Ye, Guoliang Kang, Yi Yang, and Jianbin Jiao. Image-image domain adaptation with preserved self-similarity and domain-dissimilarity for person re-identification. InProceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018. 2

    work page 2018

  6. [6]

    An image is worth 16x16 words: Transformers for image recognition at scale

    Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Syl- vain Gelly, Jakob Uszkoreit, and Neil Houlsby. An image is worth 16x16 words: Transformers for image recognition at scale. InInternational Conference on Learning Representa- tions (ICLR), 2021. 5

    work page 2021

  7. [7]

    Unsupervised person re-identification: Clustering and fine- tuning.ACM Transactions on Multimedia Computing, Com- munications, and Applications (TOMM), 14(4):1–18, 2018

    Hehe Fan, Liang Zheng, Chenggang Yan, and Yi Yang. Unsupervised person re-identification: Clustering and fine- tuning.ACM Transactions on Multimedia Computing, Com- munications, and Applications (TOMM), 14(4):1–18, 2018. 2, 3, 5, 7

    work page 2018

  8. [8]

    Mixture-of-loras: An efficient multitask tuning method for large language models

    Wenfeng Feng, Chuzhan Hao, Yuewei Zhang, Yu Han, and Hao Wang. Mixture-of-loras: An efficient multitask tuning method for large language models. InProceedings of the Joint International Conference on Computational Linguis- tics, Language Resources and Evaluation (LREC-COLING), pages 11371–11380, 2024. 3

    work page 2024

Show all 56 references
  1. [9]

    Unsupervised domain adaptation by backpropagation

    Yaroslav Ganin and Victor Lempitsky. Unsupervised domain adaptation by backpropagation. InProceedings of the In- ternational Conference on Machine Learning (ICML), pages 1180–1189, 2015. 2

    2015

  2. [10]

    Task singular vectors: Reducing task interference in model merging

    Antonio Andrea Gargiulo, Donato Crisostomi, Maria Sofia Bucarelli, Simone Scardapane, Fabrizio Silvestri, and Emanuele Rodol `a. Task singular vectors: Reducing task interference in model merging. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognit...

    2025

  3. [11]

    Mutual mean- teaching: Pseudo label refinery for unsupervised domain adaptation on person re-identification

    Yixiao Ge, Dapeng Chen, and Hongsheng Li. Mutual mean- teaching: Pseudo label refinery for unsupervised domain adaptation on person re-identification. InInternational Con- ference on Learning Representations (ICLR), 2020. 1, 5, 7

    2020

  4. [12]

    Self-paced contrastive learning with hybrid mem- ory for domain adaptive object re-id

    Yixiao Ge, Feng Zhu, Dapeng Chen, Rui Zhao, and hong- sheng Li. Self-paced contrastive learning with hybrid mem- ory for domain adaptive object re-id. InAdvances in Neural Information Processing Systems (NeurIPS), 2020. 1, 2, 5, 7

    2020

  5. [13]

    Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio

    Ian J. Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. Generative adversarial nets. InAdvances in Neural Information Processing Systems (NeurIPS), pages 2672–2680, 2014. 2

    2014

  6. [14]

    Source-free domain adaptation of weakly- supervised object localization models for histology

    Alexis Guichemerre, Soufiane Belharbi, Tsiry Mayet, Sha- keeb Murtaza, Pourya Shamsolmoali, Luke Mccaffrey, and Eric Granger. Source-free domain adaptation of weakly- supervised object localization models for histology. InPro- ceedings of the IEEE/CVF Conference on Computer Vi...

  7. [15]

    Transreid: Transformer-based object re- identification

    Shuting He, Hao Luo, Pichao Wang, Fan Wang, Hao Li, and Wei Jiang. Transreid: Transformer-based object re- identification. InProceedings of the IEEE/CVF Inter- national Conference on Computer Vision (ICCV), pages 15013–15022, 2021. 1, 5

    2021

  8. [16]

    Secret: Self-consistent pseudo label refinement for unsupervised domain adaptive person re-identification

    Tao He, Leqi Shen, Yuchen Guo, Guiguang Ding, and Zhen- hua Guo. Secret: Self-consistent pseudo label refinement for unsupervised domain adaptive person re-identification. In Proceedings of the AAAI conference on Artificial Intelligence (AAAI), pages 879–887, 2022. 1, 2

    2022

  9. [17]

    Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen- Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen

    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. InPro- ceedings of the International Conference on Learning Rep- resentations (ICLR), 2022. 2, 3

    2022

  10. [18]

    Lorahub: Efficient cross-task gener- alization via dynamic loRA composition

    Chengsong Huang, Qian Liu, Bill Yuchen Lin, Tianyu Pang, Chao Du, and Min Lin. Lorahub: Efficient cross-task gener- alization via dynamic loRA composition. InConference on Language Modeling (COLM), 2024. 3, 8

    2024

  11. [19]

    Editing models with task arithmetic.arXiv preprint arXiv:2212.04089, 2022

    Gabriel Ilharco, Marco Tulio Ribeiro, Mitchell Wortsman, Suchin Gururangan, Ludwig Schmidt, Hannaneh Hajishirzi, and Ali Farhadi. Editing models with task arithmetic.arXiv preprint arXiv:2212.04089, 2022. 8

    2022 arXiv

  12. [20]

    Towards discriminative representa- tion learning for unsupervised person re-identification

    Takashi Isobe, Dong Li, Lu Tian, Weihua Chen, Yi Shan, and Shengjin Wang. Towards discriminative representa- tion learning for unsupervised person re-identification. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pages 8526–8536, 2021. 1, 2

    2021

  13. [21]

    Samee Ullah Khan, Tanveer Hussain, Amin Ullah, and Sung Wook Baik. Deep-reid: Deep features and autoencoder assisted image patching strategy for person re-identification in smart cities surveillance.Multimedia Tools and Applica- tions, 83(5):15079–15100, 2024. 1

    2024

  14. [22]

    On convergence and stability of gans.arXiv preprint arXiv:1705.07215, 2017

    Naveen Kodali, Jacob Abernethy, James Hays, and Zsolt Kira. On convergence and stability of gans.arXiv preprint arXiv:1705.07215, 2017. 2

    2017 arXiv

  15. [23]

    Similarity of neural network represen- tations revisited

    Simon Kornblith, Mohammad Norouzi, Honglak Lee, and Geoffrey Hinton. Similarity of neural network represen- tations revisited. InInternational conference on machine learning, pages 3519–3529. PMlR, 2019. 11

    2019

  16. [24]

    Camera-driven rep- resentation learning for unsupervised domain adaptive per- son re-identification

    Geon Lee, Sanghoon Lee, Dohyung Kim, Younghoon Shin, Yongsang Yoon, and Bumsub Ham. Camera-driven rep- resentation learning for unsupervised domain adaptive per- son re-identification. InProceedings of the IEEE/CVF In- ternational Conference on Computer Vision (ICCV), pages 11...

    2023

  17. [25]

    Combined depth space based architecture search for person re-identification

    Hanjun Li, Gaojie Wu, and Wei-Shi Zheng. Combined depth space based architecture search for person re-identification. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 6729–6738,

  18. [26]

    Deep- reid: Deep filter pairing neural network for person re- identification

    Wei Li, Rui Zhao, Tong Xiao, and Xiaogang Wang. Deep- reid: Deep filter pairing neural network for person re- identification. InProceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2014. 5, 9, 10

    2014

  19. [27]

    Reliability exploration with self-ensemble learning for domain adaptive person re- identification

    Zongyi Li, Yuxuan Shi, Hefei Ling, Jiazhong Chen, Qian Wang, and Fengfan Zhou. Reliability exploration with self-ensemble learning for domain adaptive person re- identification. InProceedings of the AAAI conference on Artificial Intelligence (AAAI), pages 1527–1535, 2022. 1, 2

    2022

  20. [28]

    Moelora: Contrastive learning guided mixture of experts on parameter-efficient fine-tuning for large language models.arXiv preprint arXiv:2402.12851,

    Tongxu Luo, Jiahe Lei, Fangyu Lei, Weihao Liu, Shizhu He, Jun Zhao, and Kang Liu. Moelora: Contrastive learning guided mixture of experts on parameter-efficient fine-tuning for large language models.arXiv preprint arXiv:2402.12851,

    arXiv

  21. [29]

    Pose-guided feature alignment for occluded person re-identification

    Jiaxu Miao, Yu Wu, Ping Liu, Yuhang Ding, and Yi Yang. Pose-guided feature alignment for occluded person re-identification. InProceedings of the IEEE/CVF Interna- tional Conference on Computer Vision (ICCV), 2019. 9, 10, 11

    2019

  22. [30]

    A culturally-aware benchmark for person re-identification in modest attire.Engineering Ap- plications of Artificial Intelligence, 158:111494, 2025

    Alireza Sedighi Moghaddam, Fatemeh Anvari, Mohammad- javad Mirshekari Haghighi, Mohammadali Fakhari, and Mo- hammad Reza Mohammadi. A culturally-aware benchmark for person re-identification in modest attire.Engineering Ap- plications of Artificial Intelligence, 158:111494, 202...

    2025

  23. [31]

    Learning to route among specialized experts for zero- shot generalization

    Mohammed Muqeeth, Haokun Liu, Yufan Liu, and Colin Raffel. Learning to route among specialized experts for zero- shot generalization. InProceedings of the International Con- ference on Machine Learning (ICML), pages 1496–1513,

  24. [32]

    Moment matching for multi-source domain adaptation

    Xingchao Peng, Qinxun Bai, Xide Xia, Zijun Huang, Kate Saenko, and Bo Wang. Moment matching for multi-source domain adaptation. InProceedings of the IEEE/CVF Inter- national Conference on Computer Vision (ICCV), 2019. 2

    2019

  25. [33]

    Performance measures and a data set for multi-target, multi-camera tracking

    Ergys Ristani, Francesco Solera, Roger Zou, Rita Cucchiara, and Carlo Tomasi. Performance measures and a data set for multi-target, multi-camera tracking. InProceedings of the European Conference on Computer Vision (ECCV) Work- shops, pages 17–35, 2016. 5, 9, 10

    2016

  26. [34]

    Model merging with svd to tie the knots.arXiv preprint arXiv:2410.19735, 2024

    George Stoica, Pratik Ramesh, Boglarka Ecsedi, Leshem Choshen, and Judy Hoffman. Model merging with svd to tie the knots.arXiv preprint arXiv:2410.19735, 2024. 8

    2024 arXiv

  27. [35]

    Manifold mixup: Better representations by interpolating hidden states

    Vikas Verma, Alex Lamb, Christopher Beckham, Amir Na- jafi, Ioannis Mitliagkas, David Lopez-Paz, and Yoshua Ben- gio. Manifold mixup: Better representations by interpolating hidden states. InProceedings of the International Confer- ence on Machine Learning (ICML), pages 6438–6...

    2019

  28. [36]

    Person transfer gan to bridge domain gap for person re- identification

    Longhui Wei, Shiliang Zhang, Wen Gao, and Qi Tian. Person transfer gan to bridge domain gap for person re- identification. InProceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018. 1, 2, 5, 9, 10

    2018

  29. [37]

    Identifying unknown instances for au- tonomous driving

    Kelvin Wong, Shenlong Wang, Mengye Ren, Ming Liang, and Raquel Urtasun. Identifying unknown instances for au- tonomous driving. InProceedings of the Conference on Robot Learning (CoRL), pages 384–393, 2020. 1

    2020

  30. [38]

    Mixture of lora experts

    Xun Wu, Shaohan Huang, and Furu Wei. Mixture of lora experts. InInternational Conference on Learning Represen- tations (ICLR), 2024. 3, 5, 8, 9, 10

    2024

  31. [39]

    Distilling consistent relations for multi-source domain adap- tive person re-identification.Pattern Recognition (PR), 157: 110821, 2025

    Yuqiao Xian, Yi-Xing Peng, Xing Sun, and Wei-Shi Zheng. Distilling consistent relations for multi-source domain adap- tive person re-identification.Pattern Recognition (PR), 157: 110821, 2025. 2, 3, 5, 7

    2025

  32. [40]

    Ties-merging: Resolving interference when merging models

    Prateek Yadav, Derek Tam, Leshem Choshen, Colin A Raf- fel, and Mohit Bansal. Ties-merging: Resolving interference when merging models. InAdvances in Neural Information Processing Systems, pages 7093–7115. Curran Associates, Inc., 2023. 8

    2023

  33. [41]

    Ties-merging: Resolving interference when merging models.Advances in Neural Information Pro- cessing Systems, 36:7093–7115, 2023

    Prateek Yadav, Derek Tam, Leshem Choshen, Colin A Raf- fel, and Mohit Bansal. Ties-merging: Resolving interference when merging models.Advances in Neural Information Pro- cessing Systems, 36:7093–7115, 2023. 3

    2023

  34. [42]

    Cheb-gr: Re- thinking k-nearest neighbor search in re-ranking for person re-identification

    Jinxi Yang, He Li, Bo Du, and Mang Ye. Cheb-gr: Re- thinking k-nearest neighbor search in re-ranking for person re-identification. InProceedings of the Computer Vision and Pattern Recognition Conference (CVPR), pages 19261– 19270, 2025. 1

    2025

  35. [43]

    Curriculum manager for source selection in multi- source domain adaptation

    Luyu Yang, Yogesh Balaji, Ser-Nam Lim, and Abhinav Shri- vastava. Curriculum manager for source selection in multi- source domain adaptation. InProceedings of the European Conference on Computer Vision (ECCV), pages 608–624,

  36. [44]

    Deep learning for person re- identification: A survey and outlook.IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 44(6): 2872–2893, 2021

    Mang Ye, Jianbing Shen, Gaojie Lin, Tao Xiang, Ling Shao, and Steven CH Hoi. Deep learning for person re- identification: A survey and outlook.IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 44(6): 2872–2893, 2021. 1

    2021

  37. [45]

    Language models are super mario: Absorbing abilities from homologous models as a free lunch

    Le Yu, Bowen Yu, Haiyang Yu, Fei Huang, and Yongbin Li. Language models are super mario: Absorbing abilities from homologous models as a free lunch. InProceedings of the In- ternational Conference on Machine Learning (ICML), 2024. 3

    2024

  38. [46]

    Pha: Patch-wise high-frequency augmentation for transformer-based person re-identification

    Guiwei Zhang, Yongfei Zhang, Tianyu Zhang, Bo Li, and Shiliang Pu. Pha: Patch-wise high-frequency augmentation for transformer-based person re-identification. InProceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 14133–14142, 2023. 1

    2023

  39. [47]

    Seeing like a human: Asynchronous learning with dynamic progressive refinement for person re-identification

    Quan Zhang, Jianhuang Lai, Zhanxiang Feng, and Xiao- hua Xie. Seeing like a human: Asynchronous learning with dynamic progressive refinement for person re-identification. IEEE Transactions on Image Processing (TIP), 31:352–365,

  40. [48]

    Unsupervised domain adap- tation with noise resistible mutual-training for person re- identification

    Fang Zhao, Shengcai Liao, Guo-Sen Xie, Jian Zhao, Kai- hao Zhang, and Ling Shao. Unsupervised domain adap- tation with noise resistible mutual-training for person re- identification. InProceedings of the European Conference on Computer Vision (ECCV), pages 526–544, 2020. 1, 2

    2020

  41. [49]

    Exploiting sample uncertainty for domain adaptive person re-identification

    Kecheng Zheng, Cuiling Lan, Wenjun Zeng, Zhizheng Zhang, and Zheng-Jun Zha. Exploiting sample uncertainty for domain adaptive person re-identification. InProceed- ings of the AAAI conference on Artificial Intelligence (AAAI), pages 3538–3546, 2021. 1, 2, 5, 7

    2021

  42. [50]

    Group-aware label transfer for do- main adaptive person re-identification

    Kecheng Zheng, Wu Liu, Lingxiao He, Tao Mei, Jiebo Luo, and Zheng-Jun Zha. Group-aware label transfer for do- main adaptive person re-identification. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 5310–5319, 2021. 1

    2021

  43. [51]

    Group-aware label transfer for do- main adaptive person re-identification

    Kecheng Zheng, Wu Liu, Lingxiao He, Tao Mei, Jiebo Luo, and Zheng-Jun Zha. Group-aware label transfer for do- main adaptive person re-identification. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 5310–5319, 2021. 5, 7

    2021

  44. [52]

    Scalable person re-identification: A benchmark

    Liang Zheng, Liyue Shen, Lu Tian, Shengjin Wang, Jing- dong Wang, and Qi Tian. Scalable person re-identification: A benchmark. InProceedings of the IEEE/CVF Interna- tional Conference on Computer Vision (ICCV), 2015. 5, 9, 10

    2015

  45. [53]

    On- line pseudo label generation by hierarchical cluster dynam- ics for adaptive person re-identification

    Yi Zheng, Shixiang Tang, Guolong Teng, Yixiao Ge, Kai- jian Liu, Jing Qin, Donglian Qi, and Dapeng Chen. On- line pseudo label generation by hierarchical cluster dynam- ics for adaptive person re-identification. InProceedings of the IEEE/CVF International Conference on Compute...

    2021

  46. [54]

    On- line pseudo label generation by hierarchical cluster dynam- ics for adaptive person re-identification

    Yi Zheng, Shixiang Tang, Guolong Teng, Yixiao Ge, Kai- jian Liu, Jing Qin, Donglian Qi, and Dapeng Chen. On- line pseudo label generation by hierarchical cluster dynam- ics for adaptive person re-identification. InProceedings of the IEEE/CVF International Conference on Compute...

    2021

  47. [55]

    Random erasing data augmentation

    Zhun Zhong, Liang Zheng, Guoliang Kang, Shaozi Li, and Yi Yang. Random erasing data augmentation. InProceedings of the AAAI Conference on Artificial Intelligence (AAAI), pages 13001–13008, 2020. 5

    2020

  48. [56]

    Dual cross-attention learning for fine-grained visual categorization and object re-identification

    Haowei Zhu, Wenjing Ke, Dong Li, Ji Liu, Lu Tian, and Yi Shan. Dual cross-attention learning for fine-grained visual categorization and object re-identification. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 4692–4702, 2022. 7

    2022

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.