Pith. sign in

REVIEW 4 major objections 5 minor 58 references

Robust Semi-Supervised Learning in Open Environments

T0 review · 4 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 Semi-supervised learning fails when unlabeled data disagree with labeled data, but robust SSL now has a taxonomy, benchmarks, and curve-based metrics to address it.

desk verdict A useful but sloppy review: the taxonomy is clean and the pointers are helpful, yet the core evaluation metric formula is mathematically broken and the inconsistency parameter t is never operationalized, so take the benchmark-and-metrics pitch with salt. read the letter →

arxiv 2412.18256 v1 pith:CULNB3DO submitted 2024-12-24 cs.LG cs.AI

Lan-Zhe Guo , Lin-Han Jia , Jie-Jing Shao , Yu-Feng Li This is my paper · ORCID
classification cs.LGcs.AI
keywords semi-supervisedlearningopenenvironmentslabelinconsistencyfeaturedistributionrobustnessanalysiscurvesafeevaluationbenchmark
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 review establishes that semi-supervised learning (SSL), which uses unlabeled data to improve models when labels are scarce, breaks down in open environments where the unlabeled data no longer matches the labeled data. The paper argues that this breakdown is a distinct and tractable research problem, and organizes the field into three types of inconsistency: label space, feature space, and data distribution. It presents a benchmark, a family of robustness metrics built on the accuracy curve $\mathrm{Acc}(t)$, and an open-source toolkit as the infrastructure for measuring progress. The practical stake is that robust SSL should improve performance in the usual case while never falling below the plain supervised baseline when unlabeled data is unreliable.

What carries the argument

The organizing device is a tripartite taxonomy—label inconsistency, feature inconsistency, and distribution inconsistency—together with the inconsistency level $t\in[0,1]$ and the accuracy function $\mathrm{Acc}(t)$. This machinery carries the argument because every method reviewed is positioned as addressing one of the three inconsistencies, and every evaluation is framed as measuring robustness across $t$ through the RAC and its derived metrics. The taxonomy and the curve turn scattered failure reports and heuristic fixes into a coherent research agenda with a shared evaluation standard.

What would settle it

If an experimenter builds a benchmark with an inconsistency that does not fit the three categories—for example, structured label noise combined with confounded features—and finds that the RAC-based metrics rank methods differently than end-to-end deployment performance does, the paper's claim that these metrics fairly capture robustness would be shown incomplete.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central claim is that open-environment SSL can and should be studied as a distinct problem, defined by an inconsistency degree $t$ between labeled and unlabeled data in one of three factors: class labels, features, or distributions. The paper asserts that existing methods can reduce the negative impact of inconsistent unlabeled data, and that evaluation must be curve-based rather than point-based: the Robustness Analysis Curve $\mathrm{Acc}(t)$ maps inconsistency level to accuracy, and metrics such as AUC, Expected Accuracy, Worst-Case Accuracy, EVM, VS, and RCC summarize robustness across all $t$. It distinguishes this problem from OOD detection and domain adaptation by the scarcity of labels in SSL, and frames safe SSL—never worse than the supervised baseline—as the target.

Load-bearing premise

The entire framework presumes that open-environment failures decompose into label, feature, and distribution inconsistency, and that a single scalar $t$ can measure each; if real failures mix these types or $t$ cannot be set operationally, the taxonomy and the curve-based metrics lose their grounding.

Editorial extensions

If this is right

  • Evaluation of SSL should report robustness across inconsistency levels, not just accuracy at $t=0$.
  • The released benchmark lets different robust SSL methods be compared on the same label, feature, and distribution inconsistencies.
  • Methods tailored to one inconsistency type, such as unseen classes, are not automatically safe for another, such as missing features, so each category needs dedicated study.
  • The safe-SSL goal sets a concrete minimum requirement: exploiting unlabeled data must not hurt relative to supervised learning.
  • Open problems identified by the paper include theoretical guarantees, tabular data, pre-trained model integration, and decision-making tasks.

Reading between the lines

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

  • If the taxonomy is right, the natural next step is a unified method that handles label, feature, and distribution inconsistency simultaneously, since the paper notes these typically co-occur.
  • The RAC-based metrics could be borrowed as a general tool for benchmarking any learner under a controllable distribution shift, not just SSL.
  • The emphasis on tabular data suggests a testable consequence: consistency regularization, which drives image-based SSL, needs a replacement for augmentation when features are heterogeneous.
  • Reading the safe-SSL objective as a formal property, one could test whether a method satisfies 'never below the supervised baseline' as a constraint rather than an aspiration.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. This review article surveys robust semi-supervised learning (SSL) in open environments, where unlabeled data may be inconsistent with labeled data in label space, feature space, or data distribution. It organizes recent methods into these three categories, describes a benchmark suite and a set of robustness-oriented performance metrics (RAC, AUC, EA, WA, EVM, VS, RCC), introduces the open-source LAMDA-SSL toolkit, and concludes with open challenges such as theoretical guarantees, tabular data, pre-trained models, and decision-making tasks.

Significance. Robust SSL in open environments is an important and growing area, and this survey provides a useful tripartite taxonomy (label, feature, distribution inconsistency) that helps structure the literature. The paper also serves a practical role: it points readers to a concrete benchmark ([18]) and an open-source toolkit (LAMDA-SSL, [51]), which are valuable community resources. The strongest contribution is the organization and the identification of under-studied directions, especially tabular data and decision-making settings. However, the paper's claim that the field now has well-defined evaluation infrastructure is weakened by the undefined inconsistency level t used by all the proposed metrics and by a concrete mathematical error in the RCC formula in Table 1; these issues must be fixed before the survey can be relied on as a guide to evaluating robust SSL methods.

major comments (4)
  1. [§2 and §6.2] The inconsistency level t is not operationally defined for feature inconsistency and distribution inconsistency. In §2, t is introduced only as 'the degree of inconsistency' in [0,1] with higher values meaning more inconsistent unlabeled instances. For label inconsistency, t could naturally be the fraction of unlabeled examples from unseen classes, but the benchmark constructions in §6.1 (random feature masking, grayscale conversion, text truncation, centroid-distance filtering, Image-CLEF/Amazon shifts) do not specify how each construction maps to a continuous t. Since Acc(t) and every derived metric in §6.2 (RAC, AUC, EA, WA, EVM, VS, RCC) depend on this mapping, the evaluation framework is underspecified as stated and cannot be applied consistently across datasets or across the three inconsistency types. Please state the operational definition of t for each construction or, if t is defined in [18], reproduce the definition here.
  2. [Table 1] The Robust Correlation Coefficient formula is mathematically invalid as printed. Since ∫_0^1 t^2 dt = 1/3, the denominator contains sqrt(1/3 − 1) = sqrt(−2/3), which is imaginary over the reals. The numerator is also not a centered covariance: a correlation between t and Acc(t) requires subtracting (1/2)∫_0^1 Acc(t) dt from the first integral. If RCC is meant to be a Pearson-type correlation, the correct form is [∫_0^1 t·Acc(t) dt − (1/2)∫_0^1 Acc(t) dt] divided by sqrt(∫_0^1 t^2 dt − 1/4) times sqrt(∫_0^1 Acc^2(t) dt − (∫_0^1 Acc(t) dt)^2). The printed formula must be corrected and every displayed formula in Table 1 checked for similar typesetting or derivation errors, since these metrics are the paper's proposed evaluation standard.
  3. [§5 and References [35], [36]] Two citations appear to be misassigned. Reference [35] (Zhao et al., 'Adaptivity and non-stationarity: Problem-dependent dynamic regret for online convex optimization') is cited in §5 to support the example that unlabeled image data may contain cartoon images; that reference is about online convex optimization, not about image distribution shifts. Reference [36] (Mo et al., S-CLIP, a semi-supervised vision-language learning method) is cited in §5 for remote sensing scenarios and again in §7 for vision-language models; the S-CLIP paper does not appear to address remote sensing. Please replace these with citations that actually support the respective claims, or remove them.
  4. [§6.1–§6.3] The benchmark, the performance measures, and the toolkit are all drawn from the same research group's work ([18], [42], [51]). The text presents these as established standards, but it does not mention their provenance or any independent validation. For a review article whose central claim is that the community now has benchmarks and metrics for robust SSL, this is a notable gap. At minimum, state that these resources were introduced by the authors' own group and, if possible, cite independent applications or comparative evaluations that use them.
minor comments (5)
  1. [§6.2] The acronym RAC is used in the text but is never defined in the main text or in Table 1; please spell out 'Robustness Analysis Curve' at first use and ensure the acronym is consistent with its definition in §6.2.
  2. [Table 1] The notation PT(t) appears in the caption but is not defined before the table; please define PT as the distribution over inconsistency levels, and clarify whether PT is a known input or estimated from data.
  3. [References] Several references have incomplete or inconsistent bibliographic information, e.g., [22] lacks a year and page range, and [31] lacks page numbers. Please standardize the reference list.
  4. [§6.1] The phrase 'construct inconsistent labeled space' is ambiguous; it should read 'construct an inconsistent label space' or similar, and the sentence would benefit from an explicit statement of how many classes are discarded and how the remaining classes define the target task.
  5. [§8] The conclusion that 'It can hardly be a thorough review of all the relevant work and is mostly a brief review' is self-deprecating but accurate; however, the paper would be strengthened by a short paragraph in the introduction that explicitly states the selection criteria for the works surveyed.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the paper is a survey and does not derive predictions from fitted inputs or self-cited theorems.

full rationale

This is a review article, not a derivation chain. It introduces a taxonomy of open-environment SSL inconsistencies, summarizes existing methods, and presents evaluation benchmarks and metrics. No parameter is fitted and then renamed as a prediction, and no equation reduces to its input by construction. The benchmark and metrics sections rely on prior work by the same group ([18], [42], [51]), but these are presented as published, externally available artifacts with a released toolkit (LAMDA-SSL) and not used to infer new results within this paper. The central claim that inconsistent unlabeled data can degrade SSL is also supported by independent citations such as [4]. The informal definition of the inconsistency level t and the apparent typo in the RCC formula in Table 1 are correctness or presentation concerns, not circularity. Under the given criteria, there is no specific quoted step that exhibits a self-definitional reduction, a fitted-input-called-prediction, or a load-bearing self-citation chain.

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

The review introduces no new fitted parameters or postulated entities. Its groundedness depends on the three domain assumptions above, all of which are taken as given from the prior literature.

assumptions (3)
  • domain assumption Open-environment SSL problems can be decomposed into label, feature, and distribution inconsistency between labeled and unlabeled data.
    This taxonomy structures the entire survey (Sections 3-5). If other inconsistency types dominate, the review is incomplete.
  • domain assumption Exploiting inconsistent unlabeled data causes performance degradation, sometimes worse than supervised learning.
    Stated in the introduction and Section 3, citing references [3-8]. The review does not re-derive or test this empirically.
  • domain assumption Acc(t) can be defined and measured as a function of the inconsistency level t.
    The evaluation metrics in Table 1 assume a continuous, quantifiable inconsistency level t. The paper does not show such a quantification is well-defined for all three inconsistency types.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Robust Semi-Supervised Learning in Open Environments." pith.science (2026). https://pith.science/paper/CULNB3DO

@misc{pith2026241218256,
  author       = {Pith},
  title        = {Pith review of: Robust Semi-Supervised Learning in Open Environments},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CULNB3DO}},
  note         = {Machine review of arXiv:2412.18256}
}
read the original abstract

Semi-supervised learning (SSL) aims to improve performance by exploiting unlabeled data when labels are scarce. Conventional SSL studies typically assume close environments where important factors (e.g., label, feature, distribution) between labeled and unlabeled data are consistent. However, more practical tasks involve open environments where important factors between labeled and unlabeled data are inconsistent. It has been reported that exploiting inconsistent unlabeled data causes severe performance degradation, even worse than the simple supervised learning baseline. Manually verifying the quality of unlabeled data is not desirable, therefore, it is important to study robust SSL with inconsistent unlabeled data in open environments. This paper briefly introduces some advances in this line of research, focusing on techniques concerning label, feature, and data distribution inconsistency in SSL, and presents the evaluation benchmarks. Open research problems are also discussed for reference purposes.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

58 extracted references · 56 canonical work pages

  1. [35]

    Adaptivity and non-stationarity: Problem- dependent dynamic regret for online convex optimization

    Zhao P, Zhang Y J, Zhang L, Zhou Z H. Adaptivity and non-stationarity: Problem- dependent dynamic regret for online convex optimization. Journal of Machine Learning Research, 2024, 25(98): 1 – 52

    work page 2024

  2. [36]

    S-CLIP: semi-supervised vision-language learning us- ing few specialist captions

    Mo S, Kim M, Lee K, Shin J. S-CLIP: semi-supervised vision-language learning us- ing few specialist captions. In: Advances in Neural Information Processing Systems. 2023

    work page 2023

  3. [18]

    Realistic evaluation of semi-supervised learning algo- rithms in open environments

    Jia L H, Guo L Z, Zhou Z, Li Y F. Realistic evaluation of semi-supervised learning algo- rithms in open environments. In: Proceed- ings of the 12th International Conference on Learning Representations. 2024

    work page 2024

  4. [51]

    LAMDA- SSL: A comprehensive semi-supervised learning toolkit

    Jia L H, Guo L Z, Zhou Z, Li Y F. LAMDA- SSL: A comprehensive semi-supervised learning toolkit. Science China Information Science, 2024, 67(1)

    work page 2024

  5. [42]

    Bidirectional adaptation for robust semi-supervised learning with incon- sistent data distributions

    Jia L H, Guo L Z, Zhou Z, Shao J J, Xi- ang Y , Li Y F. Bidirectional adaptation for robust semi-supervised learning with incon- sistent data distributions. In: Proceedings of the 40th International Conference on Ma- chine Learning. 2023, 14886–14901

    work page 2023

  6. [1]

    USB: A unified semi-supervised learning benchmark for classification

    Wang Y , Chen H, Fan Y , Sun W, Tao R, Hou W, Wang R, Yang L, Zhou Z, Guo L Z, Qi H, Wu Z, Li Y , Nakamura S, Ye W, Savvides M, Raj B, Shinozaki T, Schiele B, Wang J, Xie X, Zhang Y . USB: A unified semi-supervised learning benchmark for classification. Ad- vances in Neural Information Processing Sys- tems, 2022, 3938–3961

    work page 2022

  7. [2]

    Open-environment machine learn- ing

    Zhou Z H. Open-environment machine learn- ing. National Science Review, 2022, 9(8): nwac123

    work page 2022

  8. [3]

    A general formulation for safely exploiting weakly supervised data

    Guo L Z, Li Y F. A general formulation for safely exploiting weakly supervised data. In: Proceedings of the 32ndAAAI conference on Artificial Intelligence. 2018, 3126–3133

    work page 2018

Show all 58 references
  1. [4]

    Realistic evaluation of deep semi-supervised learning algorithms

    Oliver A, Odena A, Ra ffel C, Cubuk E D, Goodfellow I J. Realistic evaluation of deep semi-supervised learning algorithms. In: Ad- vances in Neural Information Processing Sys- tems. 2018, 3239–3250

    2018

  2. [5]

    Safe deep semi-supervised learning for unseen-class unlabeled data

    Guo L Z, Zhang Z Y , Jiang Y , Li Y F, Zhou Z H. Safe deep semi-supervised learning for unseen-class unlabeled data. In: Proceedings of the 37th International Conference on Ma- chine Learning. 2020, 3897–3906

    2020

  3. [6]

    Towards safe weakly supervised learning

    Li Y F, Guo L Z, Zhou Z H. Towards safe weakly supervised learning. IEEE Transac- tions on Pattern Analysis and Machine Intel- ligence, 2019, 43(1): 334–346

    2019

  4. [7]

    Safe semi-supervised learning: A brief introduction

    Li Y F, Liang D M. Safe semi-supervised learning: A brief introduction. Frontiers Computer Science, 2019, 13(4): 669–676 Lan-Zhe Guo et al. Robust Semi-Supervised Learning in Open Environments 9

    2019

  5. [8]

    Dualmatch: Robust semi-supervised learning with dual-level interaction

    Wang C, Cao X F, Guo L Z, Shi Z L. Dualmatch: Robust semi-supervised learning with dual-level interaction. In: Proceedings of the Joint European Conference on Ma- chine Learning and Knowledge Discovery in Databases. 2023, 102–119

    2023

  6. [9]

    A baseline for de- tecting misclassified and out-of-distribution examples in neural networks

    Hendrycks D, Gimpel K. A baseline for de- tecting misclassified and out-of-distribution examples in neural networks. In: Proceedings of the 5th International Conference on Learn- ing Representations. 2017

    2017

  7. [10]

    ODS: Test-time adaptation in the presence of open-world data shift

    Zhou Z, Guo L Z, Jia L H, Zhang D, Li Y F. ODS: Test-time adaptation in the presence of open-world data shift. In: Proceedings of the 40th International Conference on Ma- chine Learning. 2023, 42574–42588

    2023

  8. [11]

    Decoop: Robust prompt tuning with out-of- distribution detection

    Zhou Z, Yang M, Shi J X, Guo L Z, Li Y F. Decoop: Robust prompt tuning with out-of- distribution detection. In: Proceedings of the 41st International Conference on Machine Learning. 2024

    2024

  9. [12]

    Log: Active model adaptation for label-e fficient ood generalization

    Shao J J, Guo L Z, Yang X W, Li Y F. Log: Active model adaptation for label-e fficient ood generalization. Advances in Neural In- formation Processing Systems, 2022, 11023– 11034

    2022

  10. [13]

    Recent advances in open set recognition: A survey

    Geng C, Huang S J, Chen S. Recent advances in open set recognition: A survey. IEEE Transaction on Pattern Analysis and Machine Intelligence, 2021, 43(10): 3614–3631

    2021

  11. [14]

    Open-set learn- ing under covariate shift

    Shao J J, Yang X, Guo L Z. Open-set learn- ing under covariate shift. Machine Learning, 2024, 113(4): 1643–1659

    2024

  12. [15]

    SSD: A uni- fied framework for self-supervised outlier de- tection

    Sehwag V , Chiang M, Mittal P. SSD: A uni- fied framework for self-supervised outlier de- tection. In: Proceedings of the 9th Interna- tional Conference on Learning Representa- tions. 2021

    2021

  13. [16]

    A simple framework for contrastive learning of visual representations

    Chen T, Kornblith S, Norouzi M, Hinton G E. A simple framework for contrastive learning of visual representations. In: Proceedings of the 37th International Conference on Ma- chine Learning. 2020, 1597–1607

    2020

  14. [17]

    Can irrele- vant data help semi-supervised learning, why and how? In: Proceedings of the 20th ACM International Conference on Information and Knowledge Management

    Yang H, Zhu S, King I, Lyu M R. Can irrele- vant data help semi-supervised learning, why and how? In: Proceedings of the 20th ACM International Conference on Information and Knowledge Management. 2011, 937–946

    2011

  15. [19]

    STEP: Out-of-distribution detection in the presence of limited in-distribution labeled data

    Zhou Z, Guo L Z, Cheng Z, Li Y F, Pu S. STEP: Out-of-distribution detection in the presence of limited in-distribution labeled data. In: Advances in Neural Information Processing Systems. 2021, 29168–29180

    2021

  16. [20]

    Semi- supervised learning under class distribution mismatch

    Chen Y , Zhu X, Li W, Gong S. Semi- supervised learning under class distribution mismatch. In: Proceedings of the 34th AAAI Conference on Artificial Intelligence. 2020, 3569–3576

    2020

  17. [21]

    Multi- task curriculum framework for open-set semi- supervised learning

    Yu Q, Ikami D, Irie G, Aizawa K. Multi- task curriculum framework for open-set semi- supervised learning. In: Proceedings of the 16th European Conference on Computer Vi- sion. 2020, 438–454 10 Front. Comput. Sci., 2024, 0(0): 1–12

    2020

  18. [22]

    Open- Match: Open-set semi-supervised learning with open-set consistency regularization

    Saito K, Kim D, Saenko K. Open- Match: Open-set semi-supervised learning with open-set consistency regularization. In: Advances in Neural Information Processing Systems. 25956–25967

  19. [23]

    Investigating the e ffect of novel classes in semi-supervised learning

    Peng A Y , Koh Y S, Riddle P, Pfahringer B. Investigating the e ffect of novel classes in semi-supervised learning. In: Proceed- ings of the 11th Asian Conference on Ma- chine Learning. 2019, 615–630

    2019

  20. [24]

    Trash to treasure: Har- vesting OOD data with cross-modal match- ing for open-set semi-supervised learning

    Huang J, Fang C, Chen W, Chai Z, Wei X, Wei P, Lin L, Li G. Trash to treasure: Har- vesting OOD data with cross-modal match- ing for open-set semi-supervised learning. In: Proceedings of the IEEE /CVF Interna- tional Conference on Computer Vision. 2021, 8310–8319

    2021

  21. [25]

    Open-world semi-supervised learning

    Cao K, Brbic M, Leskovec J. Open-world semi-supervised learning. In: Proceedings of the 10th International Conference on Learn- ing Representations. 2022

    2022

  22. [26]

    Robust semi-supervised learning when not all classes have labels

    Guo L Z, Zhang Y G, Wu Z F, Shao J J, Li Y F. Robust semi-supervised learning when not all classes have labels. In: Ad- vances in Neural Information Processing Sys- tems. 2022, 3305–3317

    2022

  23. [27]

    Class- incremental learning: Survey and perfor- mance evaluation on image classification

    Masana M, Liu X, Twardowski B, Menta M, Bagdanov A D, Van De Weijer J. Class- incremental learning: Survey and perfor- mance evaluation on image classification. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 45(5): 5513– 5533

    2022

  24. [28]

    Deep neural net- works and tabular data: A survey

    Borisov V , Leemann T, Seßler K, Haug J, Pawelczyk M, Kasneci G. Deep neural net- works and tabular data: A survey. IEEE Transactions on Neural Networks and Learn- ing Systems, 2024, 35(6): 7499–7519

    2024

  25. [29]

    Poisoning the unlabeled dataset of semi-supervised learning

    Carlini N. Poisoning the unlabeled dataset of semi-supervised learning. In: Proceedings of the 30th USENIX Security Symposium. 2021, 1577–1592

    2021

  26. [30]

    Dehib: Deep hidden backdoor attack on semi-supervised learning via adversarial per- turbation

    Yan Z, Li G, Tian Y , Wu J, Li S, Chen M, Poor H V . Dehib: Deep hidden backdoor attack on semi-supervised learning via adversarial per- turbation. In: Proceedings of the 35th AAAI Conference on Artificial Intelligence. 2021, 10585–10593

    2021

  27. [31]

    A uni- fied framework for data poisoning attack to graph-based semi-supervised learning

    Liu X, Si S, Zhu J, Li Y , Hsieh C. A uni- fied framework for data poisoning attack to graph-based semi-supervised learning. In: Advances in Neural Information Processing Systems. 2019, 9777–9787

    2019

  28. [32]

    Virtual adversarial training: A regularization method for supervised and semi-supervised learning

    Miyato T, Maeda S, Koyama M, Ishii S. Virtual adversarial training: A regularization method for supervised and semi-supervised learning. IEEE Transactions on Pattern Anal- ysis and Machine Intelligence, 2019, 41(8): 1979–1993

    2019

  29. [33]

    Tangent-normal ad- versarial regularization for semi-supervised learning

    Yu B, Wu J, Ma J, Zhu Z. Tangent-normal ad- versarial regularization for semi-supervised learning. In: Proceedings of the IEEE Confer- ence on Computer Vision and Pattern Recog- nition. 2019, 10676–10684

    2019

  30. [34]

    Robustness to adversarial perturbations in learning from incomplete data

    Najafi A, Maeda S, Koyama M, Miyato T. Robustness to adversarial perturbations in learning from incomplete data. In: Advances in Neural Information Processing Systems. 2019, 5542–5552 Lan-Zhe Guo et al. Robust Semi-Supervised Learning in Open Environments 11

    2019

  31. [37]

    Lawgpt: A chinese legal knowledge- enhanced large language model

    Zhou Z, Shi J, Song P, Yang X, Jin Y , Guo L, Li Y . Lawgpt: A chinese legal knowledge- enhanced large language model. CoRR, 2024, abs/2406.04614

    2024 arXiv

  32. [38]

    A survey on transfer learn- ing

    Pan S J, Yang Q. A survey on transfer learn- ing. IEEE Transactions on Knowledge and Data Engineering, 2009, 22(10): 1345–1359

    2009

  33. [39]

    Distributionally robust semi- supervised learning for people-centric sens- ing

    Chen K, Yao L, Zhang D, Chang X, Long G, Wang S. Distributionally robust semi- supervised learning for people-centric sens- ing. In: Proceedings of the 33rd AAAI Conference on Artificial Intelligence. 2019, 3321–3328

    2019

  34. [40]

    Universal semi-supervised learning

    Huang Z, Xue C, Han B, Yang J, Gong C. Universal semi-supervised learning. In: Ad- vances in Neural Information Processing Sys- tems. 2021, 26714–26725

    2021

  35. [41]

    Record: Re- source constrained semi-supervised learning under distribution shift

    Guo L Z, Zhou Z, Li Y F. Record: Re- source constrained semi-supervised learning under distribution shift. In: Proceedings of the 26th ACM SIGKDD International Con- ference on Knowledge Discovery and Data Mining. 2020, 1636–1644

    2020

  36. [43]

    Distribution aligning refinery of pseudo-label for imbalanced semi-supervised learning

    Kim J, Hur Y , Park S, Yang E, Hwang S J, Shin J. Distribution aligning refinery of pseudo-label for imbalanced semi-supervised learning. Advances in Neural Information Processing Systems, 2020, 14567–14579

    2020

  37. [44]

    Crest: A class-rebalancing self-training framework for imbalanced semi-supervised learning

    Wei C, Sohn K, Mellina C, Yuille A, Yang F. Crest: A class-rebalancing self-training framework for imbalanced semi-supervised learning. In: Proceedings of the IEEE /CVF Conference on Computer Vision and Pattern Recognition. 2021, 10857–10866

    2021

  38. [45]

    Learning from imbalanced and in- complete supervision with its application to ride-sharing liability judgment

    Guo L Z, Zhou Z, Shao J J, Zhang Q, Kuang F, Li G L, Liu Z X, Wu G, Ma N, Li Q, Li Y F. Learning from imbalanced and in- complete supervision with its application to ride-sharing liability judgment. In: Proceed- ings of the 27th ACM SIGKDD Conference on Knowledge Discovery and...

    2021

  39. [46]

    Class-imbalanced semi- supervised learning with adaptive threshold- ing

    Guo L Z, Li Y F. Class-imbalanced semi- supervised learning with adaptive threshold- ing. In: Proceedings of the 39th International Conference on Machine Learning. 2022, 8082–8094

    2022

  40. [47]

    Transfer and share: Semi-supervised learn- ing from long-tailed data

    Wei T, Liu Q Y , Shi J X, Tu W W, Guo L Z. Transfer and share: Semi-supervised learn- ing from long-tailed data. Machine Learning, 2024, 113(4): 1725–1742

    2024

  41. [48]

    Caputo B, M ¨uller H, Mart´ inez-G´omez J, Villegas M, Acar B, Patricia N, Marvasti N B, ¨Usk¨udarli S, Paredes R, Cazorla M, 12 Front. Comput. Sci., 2024, 0(0): 1–12 Garc´ia-Varea I, Morell V . ImageCLEF 2014: Overview and analysis of the results. In: Pro- ceedings of the 5th...

    2024

  42. [49]

    Hidden factors and hidden topics: Understanding rating dimen- sions with review text

    McAuley J, Leskovec J. Hidden factors and hidden topics: Understanding rating dimen- sions with review text. In: Proceedings of the 7th ACM Conference on Recommender Sys- tems. 2013, 165–172

    2013

  43. [50]

    Learning word vectors for sentiment analysis

    Maas A, Daly R E, Pham P T, Huang D, Ng A Y , Potts C. Learning word vectors for sentiment analysis. In: Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Lan- guage Technologies. 2011, 142–150

    2011

  44. [52]

    A closer look at deep learning on tabular data

    Ye H J, Liu S Y , Cai H R, Zhou Q L, Zhan D C. A closer look at deep learning on tabular data. CoRR, 2024, abs/2407.00956

    2024

  45. [53]

    Identi- fying useful learnwares for heterogeneous la- bel spaces

    Guo L Z, Zhou Z, Li Y F, Zhou Z H. Identi- fying useful learnwares for heterogeneous la- bel spaces. In: Proceedings of the 40th In- ternational Conference on Machine Learning. 2023, 12122–12131

    2023

  46. [54]

    Large language models can self- improve

    Huang J, Gu S, Hou L, Wu Y , Wang X, Yu H, Han J. Large language models can self- improve. In: Proceedings of the 2023 Confer- ence on Empirical Methods in Natural Lan- guage Processing. 2023, 1051–1068

    2023

  47. [55]

    How to leverage unlabeled data in o ffline reinforcement learning

    Yu T, Kumar A, Chebotar Y , Hausman K, Finn C, Levine S. How to leverage unlabeled data in o ffline reinforcement learning. In: Proceedings of the 39th International Confer- ence on Machine Learning. 2022, 25611– 25635

    2022

  48. [56]

    O ffline imita- tion learning with model-based reverse aug- mentation

    Shao J, Shi H, Guo L, Li Y . O ffline imita- tion learning with model-based reverse aug- mentation. In: Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discov- ery and Data Mining. 2024

    2024

  49. [57]

    Semi-supervised offline reinforcement learn- ing with action-free trajectories

    Zheng Q, Hena ff M, Amos B, Grover A. Semi-supervised offline reinforcement learn- ing with action-free trajectories. In: Proceed- ings of the 40th International conference on machine learning. 2023, 42339–42362

    2023

  50. [58]

    Imitation learning from imperfection: Theoretical justi- fications and algorithms

    Li Z, Xu T, Qin Z, Yu Y , Luo Z Q. Imitation learning from imperfection: Theoretical justi- fications and algorithms. Advances in Neural Information Processing Systems, 2023

    2023

Pith tools

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