Pith. sign in

REVIEW 3 major objections 6 minor 1 cited by

Certainty and Uncertainty Guided Active Domain Adaptation

T0 review · 3 major / 6 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 Active domain adaptation improves when the model's most confident target predictions are treated as labeled data before uncertainty-based sampling.

desk verdict New combination of pseudo-label certain sampling with GP-based uncertainty querying shows real gains in active domain adaptation, but the paper needs code, error bars, a corrected table, and a measurement of pseudo-label accuracy before the claims are fully supported. read the letter →

arxiv 2505.19421 v1 pith:DZB77GBJ submitted 2025-05-26 cs.CV

Bardia Safaei , Vibashan VS , Vishal M. Patel This is my paper · ORCID
classification cs.CV
keywords activedomainadaptationlearningGaussianprocesspseudo-labelinguncertaintysamplingsampleselectionconfidence-based
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 proposes a new recipe for active domain adaptation, the setting where a model trained on labeled source data is adapted to an unlabeled target domain by querying labels for a few target samples. The paper's central claim is that the model's most confident target predictions are often correct and should be treated as labeled data before uncertainty-based sampling begins. It combines a Gaussian Process-based Active Sampling (GPAS) step, which queries the most uncertain target samples, with a Pseudo-Label-based Certain Sampling (PLCS) step, which adds the top-confidence predictions per class to the labeled set without human annotation. The paper reports that this collaborative procedure beats prior ADA methods on Office-Home (76.2% average accuracy vs 73.1% for the strongest baseline) and DomainNet (85.1% vs 82.9%), while also cutting query time. If correct, it means that confidence, not just uncertainty, is a usable signal in active adaptation.

What carries the argument

The central object is a class-wise Gaussian Process built over the feature extractor's outputs. For each predicted class $c$, the labeled features $F^l_{[c]}$ and unlabeled features $F^u_{[c]}$ define a GP with a linear kernel $K(P,Q)_{j,k} = P_j Q_k^T / (\|P_j\| \|Q_k\|)$, and the posterior covariance $\Sigma^{u,c}$ yields per-sample variances; the $b$ samples with largest $\mathrm{diag}(\Sigma^{u,c})$ are queried. The complementary object is PLCS, which takes the top $\kappa\%$ highest-probability predictions per class as pseudo-labels and moves them into the labeled set. The mechanism works as a loop: PLCS removes confident samples from the search space, GPAS selects uncertain samples that reduce the domain gap, and the growing share of confident samples feeds back into both components.

What would settle it

Measure the accuracy of the top-$\kappa\%$ confident predictions per class during the first few adaptation rounds on Office-Home and DomainNet and compare it with the gains attributed to PLCS in the ablation tables; if pseudo-label accuracy is near random while gains persist, the proposed mechanism is not what the paper describes. A cleaner experiment is to run the full method with oracle ground-truth labels for the same PLCS-selected samples and check whether the reported advantage over the pseudo-label version is large.

Watch

Extended reading notes

Core claim

The discovery the paper advances is that the search space for active domain adaptation shrinks if the model's most confident target predictions are incorporated into the labeled set before any active query is made. In each sampling round, PLCS ranks target samples by the maximum classifier probability within each predicted class and moves the top $\kappa\%$ into the labeled target set; GPAS then fits a class-wise Gaussian Process to the extracted features of labeled and unlabeled samples and selects the $b$ unlabeled samples with the highest posterior variance for human annotation. These two phases are interleaved with an Uncertainty-balanced Class Sampling step that weights target samples by their class-level GP variance and with an entropy-consistency objective on augmented target views. The paper shows empirically that each ingredient contributes to the final accuracy, and that adding PLCS to established ADA baselines improves their results as well. The headline evidence is that the full method reaches 76.2% average accuracy on Office-Home and 85.1% on DomainNet under a 5% annotation budget, surpassing previous ADA methods.

Load-bearing premise

The load-bearing premise is that the model's most confident target predictions are usually correct, because PLCS trains on them as ground truth; if a large share of those pseudo-labels are wrong, the method injects errors into the labeled set and the class-wise GP variance estimates that drive sampling become unreliable.

Editorial extensions

If this is right

  • Under the reported protocol (1% query per round, 5% total budget), the method outperforms entropy, CLUE, SDM-AG, and other baselines on both Office-Home and DomainNet.
  • Ablations in the paper indicate that replacing GPAS with random sampling, entropy, CLUE, or SDM-AG while keeping PLCS and the consistency objective yields lower accuracy, locating the benefit in the GP variance query.
  • Adding PLCS to existing ADA methods (random, entropy, CLUE, SDM-AG) improves each of them on Office-Home, implying the confident-sampling idea is a transferable augmentation rather than a quirk of one query function.
  • The certain sampling rate has an optimum: raising $\kappa$ beyond 15% eventually hurts accuracy as false pseudo-labels appear, so the benefit is bounded by pseudo-label correctness.
  • Query time stays competitive (e.g., 133.3s vs 151.4s for entropy on DomainNet R→P), which matters for large-scale deployment.

Reading between the lines

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

  • A direct test of the underlying premise would be to measure pseudo-label accuracy on the top-$\kappa\%$ confident samples; if accuracy falls sharply on a harder domain shift, the 15% certain sampling rate would need to be lowered or screened by a confidence threshold.
  • The confidence-first idea could transfer to other label-efficient settings, such as fine-tuning on a target distribution where a model's confident predictions are pre-committed before selecting uncertain examples.
  • Class-wise GP variance depends on the pseudo-label grouping; on imbalanced target classes, the per-class uncertainty estimates may need a per-class budget adjustment rather than a global top-$\kappa\%$.
  • Because the benefit of PLCS relies on confidence calibration, test-time calibration methods could change the optimal $\kappa$; one could test whether a calibrated confidence score shifts the certain-sampling schedule.
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 / 6 minor

Summary. The paper proposes a two-part active domain adaptation (ADA) method. GPAS uses class-wise Gaussian processes to estimate posterior variances of unlabeled target samples and queries the top-b uncertain samples. PLCS adds the top kappa% most confident target samples per class to the labeled set with their pseudo-labels, after which training continues with a SENTRY UDA loss and uncertainty-balanced class sampling. On Office-Home and DomainNet under a 5% annotation budget, the method reports average accuracies of 76.2% and 85.1%, outperforming SDM-AG by 3.1 and 2.2 points. Ablations show each component contributes and that PLCS can be plugged into other ADA baselines. The central claim is that using confident predictions as pseudo-labels before active sampling shrinks the search space and improves adaptation.

Significance. If the results hold, the idea is practically valuable: it offers a simple way to exploit low-cost confident pseudo-labels within an active DA loop, and the reported gains are substantial relative to prior ADA methods. The manuscript's positive features include a clear problem formulation with explicit equations, several ablations (Tables 3 and 4) that isolate the roles of GPAS, PLCS, and UCS, and a timing comparison in Table 5. However, the main claim rests on two unverified or inconsistent pieces of evidence: the accuracy of pseudo-labels is never measured, and Table 1 contains an implausible identical row for DBAL and BADGE. Since these issues concern the core empirical claim, the paper needs revision before its contribution can be considered established.

major comments (3)
  1. [Table 1] In Table 1, the DBAL and BADGE rows are identical for all 12 Office-Home transfers (e.g., 59.2, 81.0, ..., 70.6). These are distinct algorithms and cannot be expected to produce exactly matching numbers. Please correct the DBAL row or the BADGE row and re-run the comparisons; this table is the main evidence for the claimed state-of-the-art improvement.
  2. [Sec. 3.1 and Appendix 8.1] The central assumption that confident predictions 'often match the ground truth' is never quantified. PLCS treats the top-kappa% per-class predictions as ground truth in D_lt, and Eq. (3) uses the same pseudo-labels to assign unlabeled samples to class-wise GPs. If a non-negligible fraction of the top-confidence pseudo-labels are wrong, PLCS injects incorrect labels and GPAS estimates variances on incorrectly grouped samples. Appendix 8.1 mentions that false positives increase beyond a threshold but reports no false-positive rate at the operating point kappa=15%. Please add per-round pseudo-label precision, and a control experiment in which the same 15% pseudo-labeled budget is filled by random or low-confidence samples instead of top-confidence ones, to show that the gain comes from certainty rather than from extra data volume.
  3. [Sec. 4] No error bars, standard deviations, or code release are provided. Given the magnitude of the reported improvements and the identical DBAL/BADGE rows, single-run accuracies are insufficient to support the claim of consistent superiority. Please report mean +/- standard deviation over at least three runs for the main tables, or make code available so results can be reproduced.
minor comments (6)
  1. [Sec. 8.1] The first sentence of Sec. 8.1 says the active sampling rate is fixed at 15%, but the caption of Fig. 4 says it is fixed at 5%; please correct this typo.
  2. [Sec. 4] The text says the comparison includes DBAL and CoreSet, but Table 2 has no rows for either method; please add the missing rows or remove the methods from the comparison list.
  3. [Sec. 1] The Introduction claims that adding confident samples reduced selection time from 601s to 540s, but Table 5 reports only per-round times of 89.0s and 133.3s for the proposed method; please provide the exact protocol for the 601s/540s comparison or remove the claim.
  4. [Eq. (11)] The EMA coefficient alpha in Eq. (11) is never specified; please report the value used and, if possible, its sensitivity.
  5. [Eqs. (5)-(6)] The GP posterior in Eqs. (5)-(6) uses no noise term; please clarify whether the feature values F_l are treated as noiseless observations and whether the kernel inversion is intended to be a deterministic interpolation.
  6. [Algorithm 1, line 9] In Algorithm 1, line 9, the set X_conf is added to D_lt, but X_conf entries carry pseudo-labels rather than oracle labels; consider renaming D_lt or adding a comment to distinguish pseudo-labeled from oracle-labeled target data.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: GPAS and PLCS are empirically evaluated against external benchmarks, and no equation reduces to its own inputs.

full rationale

The paper's claimed gains (Tables 1–2) are measured on held-out target-domain test sets against external baselines; no component is fitted to those test labels. GPAS is a standard GP posterior-variance acquisition (Eqs. 4–8) whose inputs are labeled features and pseudo-labeled target features; selecting high-variance samples is an operational active-learning rule, not a prediction derived from the target accuracy. PLCS selects the per-class top-κ% confidence samples (Algorithm 1, lines 6–11); this is self-training, and its validity is tested by ablations (Tables 3–4) rather than assumed by construction. The κ=15% working point is chosen via the Appendix 8.1 sweep and then held fixed; it is a hyperparameter, not a renamed fitted output. Self-citations [26,27,31] appear only in the related-work survey (Sec. 8.6) and are not load-bearing; the GP machinery is attributed to standard GP references and to an external active-learning-with-GP citation. The skeptic's concern that pseudo-label accuracy is never measured is a correctness/robustness risk (the Appendix admits false positives beyond a threshold), but it does not make any equation equivalent to its inputs. There is no circular derivation chain.

Assumptions & free parameters 4 free parameters · 3 assumptions · 0 invented entities

The contribution is algorithmic; no new objects are postulated. The ledger lists the tuned hyperparameters and the reliability assumptions on pseudo-labels and GP uncertainty.

free parameters (4)
  • certain sampling rate kappa = kappa=1 initial, +1 per round, 15% cumulative
    Chosen by experiments on the evaluation datasets; Appendix 8.1 states 15% is consistently optimal across all DA datasets.
  • EMA coefficient alpha in Eq. 11 = not reported
    Hyperparameter in the uncertainty update, no value given in the paper.
  • loss weight lambda = 1
    Set to 1, not tuned.
  • active budget per round = 1% per round, 5% total
    Standard protocol, but the choice affects comparisons.
assumptions (3)
  • domain assumption High-confidence target predictions are accurate enough to be used as ground truth.
    Central to PLCS, stated in the abstract but never measured in the paper.
  • domain assumption GP posterior variance on deep features is a valid uncertainty measure.
    Used in GPAS and UCS, not theoretically justified beyond the standard GP regression framework.
  • domain assumption SENTRY loss improves target adaptation.
    The method relies on SENTRY (Eq. 10) as the UDA component, inherited from prior work.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Certainty and Uncertainty Guided Active Domain Adaptation." pith.science (2026). https://pith.science/paper/DZB77GBJ

@misc{pith2026250519421,
  author       = {Pith},
  title        = {Pith review of: Certainty and Uncertainty Guided Active Domain Adaptation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DZB77GBJ}},
  note         = {Machine review of arXiv:2505.19421}
}
read the original abstract

Active Domain Adaptation (ADA) adapts models to target domains by selectively labeling a few target samples. Existing ADA methods prioritize uncertain samples but overlook confident ones, which often match ground-truth. We find that incorporating confident predictions into the labeled set before active sampling reduces the search space and improves adaptation. To address this, we propose a collaborative framework that labels uncertain samples while treating highly confident predictions as ground truth. Our method combines Gaussian Process-based Active Sampling (GPAS) for identifying uncertain samples and Pseudo-Label-based Certain Sampling (PLCS) for confident ones, progressively enhancing adaptation. PLCS refines the search space, and GPAS reduces the domain gap, boosting the proportion of confident samples. Extensive experiments on Office-Home and DomainNet show that our approach outperforms state-of-the-art ADA methods.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

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

  1. StepAL: Step-aware Active Learning for Cataract Surgical Videos

    cs.CV 2025-07 conditional novelty 6.0 of 10

    StepAL is an active learning method that selects whole surgical videos that are both uncertain and step-diverse, improving step recognition accuracy with fewer annotations.

Reference graph

Works this paper leans on

58 extracted references · 51 canonical work pages · cited by 1 Pith paper

  1. [1]

    However, their performance degrades on data different from the training set [1]

    INTRODUCTION Deep Neural Networks (DNNs) have significantly advanced computer vision tasks. However, their performance degrades on data different from the training set [1]. To address this limitation, Unsupervised Domain Adaptation (UDA) methods [2, 3, 4] improve generalization from a labeled source to an unlabeled target domain but still lag behind fully...

    arXiv 2025

  2. [2]

    Gaussian Processes (GP) are non- parametric probabilistic models that generate uncertainty- aware predictions, making them suitable for semi-supervised and active learning [11, 12]

    BACKGROUND Gaussian processes. Gaussian Processes (GP) are non- parametric probabilistic models that generate uncertainty- aware predictions, making them suitable for semi-supervised and active learning [11, 12]. A GP f (x) is an infinite set of random variables where any finite subset follows a joint Gaussian distribution [13]. It is defined as: f (x) ∼ ...

    work page

  3. [3]

    Pseudo-Label based Certain Sampling Existing Active Learning (AL) methods focus on selecting uncertain samples while ignoring confident ones

    APPROACH 3.1. Pseudo-Label based Certain Sampling Existing Active Learning (AL) methods focus on selecting uncertain samples while ignoring confident ones. We improve this by leveraging pseudo-labels for confident target samples, reducing the need for active labeling and better approximat- ing the target distribution. This makes AL more efficient by shrin...

    work page

  4. [4]

    EXPERIMENTS AND RESULTS We conduct extensive experiments on two commonly used DA datasets, namely DomainNet [19] and Office-Home [20] to verify the effectiveness of our approach. We compare Method Office-Home A→C A→P A→R C→A C→P C→R P →A P →C P →R R→A R→C R→P Avg ResNet [15] 42.1 66.3 73.3 50.7 59.0 62.6 51.9 37.9 71.2 65.2 42.6 76.6 58.3 Random 56.8 78.0...

    work page

  5. [5]

    For the PLCS phase, we initialize κ = 1 and increase it by 1 per round, resulting in 15% certain pseudo-labels used during training

    Each sampling round is followed by 3 UDA epochs, with a five-epoch warm-up before the first round. For the PLCS phase, we initialize κ = 1 and increase it by 1 per round, resulting in 15% certain pseudo-labels used during training. Office-Home results . The classification results for the Office-Home dataset are shown in Table 1. This table shows the super...

    work page

  6. [6]

    We conduct extensive abla- tion studies to assess the effectiveness of each component in our method, reporting results in Table 3 on the Office-Home dataset

    ABLATION STUDY Effect of different components. We conduct extensive abla- tion studies to assess the effectiveness of each component in our method, reporting results in Table 3 on the Office-Home dataset. The first row presents unsupervised domain adapta- tion using SENTRY . Variant 1 applies random sampling with SENTRY loss, while Variant 2 uses only GPA...

    work page

  7. [7]

    CONCLUSION We proposed a novel collaborative framework for Active Domain Adaptation (ADA) that incorporates both uncertain and certain target samples during the training process. Our approach involves using a Gaussian Process-based Active Sampling (GPAS) strategy to identify uncertain samples and a Pseudo-Label-based Certain Sampling (PLCS) strategy to id...

    work page

  8. [8]

    A theory of learning from different domains,

    Shai Ben-David, John Blitzer, Koby Crammer, Alex Kulesza, Fernando Pereira, and Jennifer Wortman Vaughan, “A theory of learning from different domains,” Machine learning, vol. 79, pp. 151–175, 2010. 1

    work page 2010

Show all 58 references
  1. [9]

    Adversarial discriminative domain adapta- tion,

    Eric Tzeng, Judy Hoffman, Kate Saenko, and Trevor Darrell, “Adversarial discriminative domain adapta- tion,” in Proceedings of the IEEE conference on com- puter vision and pattern recognition , 2017, pp. 7167–

    2017

  2. [10]

    Cycada: Cycle-consistent adversarial domain adaptation,

    Judy Hoffman, Eric Tzeng, Taesung Park, Jun-Yan Zhu, Phillip Isola, Kate Saenko, Alexei Efros, and Trevor Darrell, “Cycada: Cycle-consistent adversarial domain adaptation,” in International conference on machine learning. Pmlr, 2018, pp. 1989–1998. 1, 10

    2018

  3. [11]

    Unsupervised pixel-level domain adaptation with generative adversar- ial networks,

    Konstantinos Bousmalis, Nathan Silberman, David Do- han, Dumitru Erhan, and Dilip Krishnan, “Unsupervised pixel-level domain adaptation with generative adversar- ial networks,” in Proceedings of the IEEE conference on computer vision and pattern recognition , 2017, pp. 3722–3731. 1

    2017

  4. [12]

    Active adversarial domain adaptation,

    Jong-Chyi Su, Yi-Hsuan Tsai, Kihyuk Sohn, Buyu Liu, Subhransu Maji, and Manmohan Chandraker, “Active adversarial domain adaptation,” in Proceedings of the IEEE/CVF Winter Conference on Applications of Com- puter Vision, 2020, pp. 739–748. 1, 4, 9, 10

    2020

  5. [13]

    Learning distinctive margin toward active domain adaptation,

    Ming Xie, Yuxi Li, Yabiao Wang, Zekun Luo, Zhenye Gan, Zhongyi Sun, Mingmin Chi, Chengjie Wang, and Pei Wang, “Learning distinctive margin toward active domain adaptation,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recogni- tion, 2022, pp. 7993–8...

    2022

  6. [14]

    Transferable query selection for active domain adaptation,

    Bo Fu, Zhangjie Cao, Jianmin Wang, and Mingsheng Long, “Transferable query selection for active domain adaptation,” in Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition , 2021, pp. 7272–7281. 1, 4, 9, 10

    2021

  7. [15]

    Active learning for domain adaptation: An energy-based approach,

    Binhui Xie, Longhui Yuan, Shuang Li, Chi Harold Liu, Xinjing Cheng, and Guoren Wang, “Active learning for domain adaptation: An energy-based approach,” in Pro- ceedings of the AAAI Conference on Artificial Intelli- gence, 2022, vol. 36, pp. 8708–8716. 1, 10

    2022

  8. [16]

    Active domain adaptation via clustering uncertainty-weighted embeddings,

    Viraj Prabhu, Arjun Chandrasekaran, Kate Saenko, and Judy Hoffman, “Active domain adaptation via clustering uncertainty-weighted embeddings,” in Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021, pp. 8505–8514. 1, 4, 5, 9, 10

    2021

  9. [17]

    1, Springer, 2006

    Carl Edward Rasmussen, Christopher KI Williams, et al., Gaussian processes for machine learning, vol. 1, Springer, 2006. 1

    2006

  10. [18]

    Syn2real transfer learning for image derain- ing using gaussian processes,

    Rajeev Yasarla, Vishwanath A Sindagi, and Vishal M Patel, “Syn2real transfer learning for image derain- ing using gaussian processes,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2020, pp. 2726–2736. 2

    2020

  11. [19]

    Active learning with gaussian processes for object categorization,

    Ashish Kapoor, Kristen Grauman, Raquel Urtasun, and Trevor Darrell, “Active learning with gaussian processes for object categorization,” in 2007 IEEE 11th interna- tional conference on computer vision . IEEE, 2007, pp. 1–8. 2

    2007

  12. [20]

    Gaussian processes in machine learning,

    Carl Edward Rasmussen and Christopher KI Williams, “Gaussian processes in machine learning,” Lecture notes in computer science , vol. 3176, pp. 63–71, 2004. 2

    2004

  13. [21]

    Sentry: Selective entropy optimization via committee consistency for unsupervised domain adap- tation,

    Viraj Prabhu, Shivam Khare, Deeksha Kartik, and Judy Hoffman, “Sentry: Selective entropy optimization via committee consistency for unsupervised domain adap- tation,” in Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021, pp. 8558–8567. 3, 10

    2021

  14. [22]

    Deep residual learning for image recognition,

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun, “Deep residual learning for image recognition,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 770–778. 4, 9

    2016

  15. [23]

    Active learning for convolutional neural networks: A core-set approach,

    Ozan Sener and Silvio Savarese, “Active learning for convolutional neural networks: A core-set approach,” arXiv preprint arXiv:1708.00489, 2017. 4, 9, 10

    2017 arXiv

  16. [24]

    Deep batch active learning by diverse, uncertain gradient lower bounds,

    Jordan T Ash, Chicheng Zhang, Akshay Krishnamurthy, John Langford, and Alekh Agarwal, “Deep batch active learning by diverse, uncertain gradient lower bounds,” arXiv preprint arXiv:1906.03671, 2019. 4, 10

    1906 arXiv

  17. [25]

    Discrepancy-based ac- tive learning for domain adaptation,

    Antoine de Mathelin, Francois Deheeger, Mathilde Mougeot, and Nicolas Vayatis, “Discrepancy-based ac- tive learning for domain adaptation,” arXiv preprint arXiv:2103.03757, 2021. 4, 9, 10

    2021 arXiv

  18. [26]

    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,” in Proceedings of the IEEE/CVF international conference on computer vision, 2019, pp. 1406–1415. 3

    2019

  19. [27]

    Deep hashing network for unsupervised domain adaptation,

    Hemanth Venkateswara, Jose Eusebio, Shayok Chakraborty, and Sethuraman Panchanathan, “Deep hashing network for unsupervised domain adaptation,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2017, pp. 5018–5027. 3

    2017

  20. [28]

    Pytorch: An imperative style, high-performance deep learning library,

    Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, Alban Desmaison, Andreas Kopf, Edward Yang, Zachary DeVito, Martin Raison, Alykhan Tejani, Sasank Chilamkurthy, Benoit Steiner, Lu ...

    2019

  21. [29]

    Imagenet: A large-scale hierarchical image database,

    Jia Deng, Wei Dong, Richard Socher, Li-Jia Li, Kai Li, and Li Fei-Fei, “Imagenet: A large-scale hierarchical image database,” in 2009 IEEE conference on computer vision and pattern recognition. Ieee, 2009, pp. 248–255. 9

    2009

  22. [30]

    An overview of gradient descent opti- mization algorithms,

    Sebastian Ruder, “An overview of gradient descent opti- mization algorithms,” arXiv preprint arXiv:1609.04747,

    arXiv

  23. [31]

    Conditional adversarial domain adaptation,

    Mingsheng Long, Zhangjie Cao, Jianmin Wang, and Michael I Jordan, “Conditional adversarial domain adaptation,” Advances in neural information processing systems, vol. 31, 2018. 9

    2018

  24. [32]

    Active learning on a budget: Opposite strate- gies suit high and low budgets,

    Guy Hacohen, Avihu Dekel, and Daphna Wein- shall, “Active learning on a budget: Opposite strate- gies suit high and low budgets,” arXiv preprint arXiv:2202.02794, 2022. 10

    2022 arXiv

  25. [33]

    Filter images first, gen- erate instructions later: Pre-instruction data selec- tion for visual instruction tuning,

    Bardia Safaei, Faizan Siddiqui, Jiacong Xu, Vishal M Patel, and Shao-Yuan Lo, “Filter images first, gen- erate instructions later: Pre-instruction data selec- tion for visual instruction tuning,” arXiv preprint arXiv:2503.07591, 2025. 10

    2025 arXiv

  26. [34]

    Active learn- ing for vision-language models,

    Bardia Safaei and Vishal M Patel, “Active learn- ing for vision-language models,” in 2025 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV). IEEE, 2025, pp. 4902–4912. 10

    2025

  27. [35]

    Active prompt learning in vision language models,

    Jihwan Bang, Sumyeong Ahn, and Jae-Gil Lee, “Active prompt learning in vision language models,” inProceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp. 27004–27014. 10

    2024

  28. [36]

    Active finetun- ing: Exploiting annotation budget in the pretraining- finetuning paradigm,

    Yichen Xie, Han Lu, Junchi Yan, Xiaokang Yang, Masayoshi Tomizuka, and Wei Zhan, “Active finetun- ing: Exploiting annotation budget in the pretraining- finetuning paradigm,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recogni- tion, 2023, pp. 23715...

    2023

  29. [37]

    La- tent structured active learning,

    Wenjie Luo, Alex Schwing, and Raquel Urtasun, “La- tent structured active learning,” Advances in Neural In- formation Processing Systems, vol. 26, 2013. 10

    2013

  30. [38]

    Entropic open-set active learning,

    Bardia Safaei, VS Vibashan, Celso M de Melo, and Vishal M Patel, “Entropic open-set active learning,” in Proceedings of the AAAI conference on artificial intelli- gence, 2024, vol. 38, pp. 4686–4694. 10

    2024

  31. [39]

    Margin based active learning,

    Maria-Florina Balcan, Andrei Broder, and Tong Zhang, “Margin based active learning,” in International Con- ference on Computational Learning Theory . Springer, 2007, pp. 35–50. 10

    2007

  32. [40]

    Batchbald: Efficient and diverse batch acquisition for deep bayesian active learning,

    Andreas Kirsch, Joost Van Amersfoort, and Yarin Gal, “Batchbald: Efficient and diverse batch acquisition for deep bayesian active learning,” Advances in neural in- formation processing systems, vol. 32, 2019. 10

    2019

  33. [41]

    Representative sampling for text classification using support vector machines,

    Zhao Xu, Kai Yu, V olker Tresp, Xiaowei Xu, and Jizhi Wang, “Representative sampling for text classification using support vector machines,” in Advances in Infor- mation Retrieval: 25th European Conference on IR Re- search, ECIR 2003, Pisa, Italy, April 14–16, 2003. Pro- ceedi...

    2003

  34. [42]

    Submodularity in data subset selection and active learning,

    Kai Wei, Rishabh Iyer, and Jeff Bilmes, “Submodularity in data subset selection and active learning,” in Interna- tional conference on machine learning . PMLR, 2015, pp. 1954–1963. 10

    2015

  35. [43]

    Unsupervised domain adaptation by backpropagation,

    Yaroslav Ganin and Victor Lempitsky, “Unsupervised domain adaptation by backpropagation,” in Interna- tional conference on machine learning . PMLR, 2015, pp. 1180–1189. 10

    2015

  36. [44]

    Semi-supervised domain adaptation via minimax entropy,

    Kuniaki Saito, Donghyun Kim, Stan Sclaroff, Trevor Darrell, and Kate Saenko, “Semi-supervised domain adaptation via minimax entropy,” in Proceedings of the IEEE/CVF international conference on computer vision, 2019, pp. 8050–8058. 10

    2019

  37. [45]

    Learning invariant representations and risks for semi-supervised domain adaptation,

    Bo Li, Yezhen Wang, Shanghang Zhang, Dongsheng Li, Kurt Keutzer, Trevor Darrell, and Han Zhao, “Learning invariant representations and risks for semi-supervised domain adaptation,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recogni- tion, 2021, pp...

    2021

  38. [46]

    Contrastive adaptation network for un- supervised domain adaptation,

    Guoliang Kang, Lu Jiang, Yi Yang, and Alexander G Hauptmann, “Contrastive adaptation network for un- supervised domain adaptation,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2019, pp. 4893–4902. 10

    2019

  39. [47]

    Mind the class weight bias: Weighted maximum mean discrepancy for unsu- pervised domain adaptation,

    Hongliang Yan, Yukang Ding, Peihua Li, Qilong Wang, Yong Xu, and Wangmeng Zuo, “Mind the class weight bias: Weighted maximum mean discrepancy for unsu- pervised domain adaptation,” in Proceedings of the IEEE conference on computer vision and pattern recog- nition, 2017, pp. 22...

    2017

  40. [48]

    Domain- adversarial training of neural networks,

    Yaroslav Ganin, Evgeniya Ustinova, Hana Ajakan, Pas- cal Germain, Hugo Larochelle, Franc ¸ois Laviolette, Mario Marchand, and Victor Lempitsky, “Domain- adversarial training of neural networks,” The journal of machine learning research, vol. 17, no. 1, pp. 2096– 2030, 2016. 10

    2016

  41. [49]

    Gradient distribution alignment certificates better adversarial domain adapta- tion,

    Zhiqiang Gao, Shufei Zhang, Kaizhu Huang, Qiufeng Wang, and Chaoliang Zhong, “Gradient distribution alignment certificates better adversarial domain adapta- tion,” in Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021, pp. 8937–8946. 10

    2021

  42. [50]

    Unsupervised domain adaptation for semantic segmen- tation via class-balanced self-training,

    Yang Zou, Zhiding Yu, BVK Kumar, and Jinsong Wang, “Unsupervised domain adaptation for semantic segmen- tation via class-balanced self-training,” in Proceedings of the European conference on computer vision (ECCV), 2018, pp. 289–305. 10

    2018

  43. [51]

    Class- imbalanced domain adaptation: an empirical odyssey,

    Shuhan Tan, Xingchao Peng, and Kate Saenko, “Class- imbalanced domain adaptation: an empirical odyssey,” in Computer Vision–ECCV 2020 Workshops: Glas- gow, UK, August 23–28, 2020, Proceedings, Part I 16 . Springer, 2020, pp. 585–602. 10

    2020

  44. [52]

    Confidence regularized self-training,

    Yang Zou, Zhiding Yu, Xiaofeng Liu, BVK Kumar, and Jinsong Wang, “Confidence regularized self-training,” in Proceedings of the IEEE/CVF International Confer- ence on Computer Vision, 2019, pp. 5982–5991. 10

    2019

  45. [53]

    Natural and adversarial error detection using in- variance to image transformations,

    Yuval Bahat, Michal Irani, and Gregory Shakhnarovich, “Natural and adversarial error detection using in- variance to image transformations,” arXiv preprint arXiv:1902.00236, 2019. 10

    1902 arXiv

  46. [54]

    Domain adaptation meets ac- tive learning,

    Piyush Rai, Avishek Saha, Hal Daum ´e III, and Suresh Venkatasubramanian, “Domain adaptation meets ac- tive learning,” in Proceedings of the NAACL HLT 2010 Workshop on Active Learning for Natural Language Processing, 2010, pp. 27–32. 10

    2010

  47. [55]

    S3vaada: Submodular subset selec- tion for virtual adversarial active domain adaptation,

    Harsh Rangwani, Arihant Jain, Sumukh K Aithal, and R Venkatesh Babu, “S3vaada: Submodular subset selec- tion for virtual adversarial active domain adaptation,” in Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021, pp. 7516–7525. 10

    2021

  48. [56]

    Varying Certain Sampling Rate Fig

    APPENDIX 8.1. Varying Certain Sampling Rate Fig. 4 illustrates the results of experiments conducted to ob- serve the impact of varying the certain sampling rate on the performance of our method, SDM-AG, and entropy sampling approaches. The experiments were carried out by fixin...

  49. [57]

    For the classifier, we ini- tialize weights using the Xavier initialization technique with no bias

    architecture as our feature extractor and a single fully- connected layer as the classifier. For the classifier, we ini- tialize weights using the Xavier initialization technique with no bias. We use SGD optimizer [23] with a learning rate of 0.002, a momentum of 0.9, and a we...

  50. [58]

    CLUE [9] uses pre- dictive entropy for uncertainty estimation and then samples from different clusters that are weighted by entropy to im- pose diversity

    selects samples that have high predictive entropy and tar- getness using a domain discriminator and performs adversar- ial domain alignment using DANN [41]. CLUE [9] uses pre- dictive entropy for uncertainty estimation and then samples from different clusters that are weighted...

Pith tools

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