Pith. sign in

REVIEW 3 major objections 3 minor 64 references

Synchronizing Task Behavior: Aligning Multiple Tasks during Test-Time Training

T0 review · 3 major / 3 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read Test-time training that learns relations between tasks outperforms existing methods on multi-task benchmarks.

desk verdict Real empirical contribution on multi-task TTT, but the theory should be corrected before publication. read the letter →

arxiv 2507.07778 v2 pith:CMXOPHXJ submitted 2025-07-10 cs.LG cs.AIcs.CV

Wooseong Jeong , Jegyeong Cho , Youngho Yoon , Kuk-Jin Yoon This is my paper · ORCID
classification cs.LGcs.AIcs.CV
keywords test-timetrainingadaptationmulti-tasklearningtaskrelationsmaskedautoencoderdomainshiftsynchronizationdenseprediction
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 claims that when a model performs several tasks at once and the input distribution shifts, standard test-time training adapts each task at a different pace and in incompatible directions, a problem it calls task unsynchronization. Its proposal, S4T, treats the relations between tasks—modeled as the conditional distribution of each task label given all task-specific latent vectors—as the transferable information, rather than any single auxiliary task. A Task Behavior Synchronizer learns these relations on the source domain by predicting each task's labels from masked versions of all task-specific features; at test time the model adapts by asking the synchronizer to reproduce the main decoder outputs. The paper reports that this synchronization improves multi-task test-time training performance on NYUD-v2, PASCAL-Context, and Taskonomy, and that its synchronization metrics positively correlate with adaptation gain. If correct, this redirects test-time training from finding a good surrogate task to preserving and transferring the dependency structure among the tasks themselves.

What carries the argument

The central object is the Task Behavior Synchronizer (TBS), a lightweight vision transformer that stacks task-specific latent vectors, masks the same random patch pattern across all tasks, and predicts each task's labels from the masked set. It is trained on the source domain with ground-truth labels alongside task-specific projection layers, and at test time its outputs are aligned to the main decoder outputs through the pseudo-label prediction loss. The masking mechanism forces the network to infer each task's output from the other tasks' features, which is what couples the tasks during adaptation; the paper argues this instantiates the assumption that $p(y_i \mid z_1, \dots, z_n)$ transfers across domains up to a domain-dependent transformation $f$.

What would settle it

Run S4T on a pair of domains in which the ordering of task relations is deliberately reversed—for example, depth and surface normal constraints agree in the source and conflict in the target—and compare the adaptation gain against the same model without any test-time adaptation. If the gain turns negative or falls below that of a single-task TTT method, Assumption 1 fails and the central claim is falsified.

Watch

Extended reading notes

Core claim

The paper's central claim is that task relations—not a single auxiliary self-supervised task—should carry test-time adaptation in multi-task settings. It identifies a failure mode it calls unsynchronized task behavior: under a domain shift, the adaptation steps that optimize one task can be suboptimal for another, so methods designed around one auxiliary task adapt tasks at different rates and to different endpoints. S4T addresses this by training a Task Behavior Synchronizer on the source domain to approximate $p(y_i \mid z_1, \dots, z_n)$, the conditional distribution of each task label given all task-specific latent vectors, using masked inputs so that each task must be inferred from the others. At test time, the method minimizes the pseudo-label prediction loss that aligns TBS predictions from masked latents with main-decoder predictions from full latents; Proposition 1 bounds the target task loss by this discrepancy plus a residual controlled by the two assumptions. The paper reports higher $\triangle_{TTT}$ than prior TTA/TTT methods across Taskonomy→NYUD-v2, Taskonomy→PASCAL-Context, NYUD-v2→Taskonomy, and PASCAL-Context→Taskonomy, and shows with step variance, DTW, and cosine similarity that stronger synchronization accompanies better multi-task performance.

Load-bearing premise

The load-bearing premise is that the dependency pattern among tasks—how each task's label relates to the latent features of the other tasks—stays approximately the same when the domain changes, up to a transformation the method never estimates; if the pattern itself shifts, the test-time pseudo-label alignment can steer the tasks toward source-biased couplings and hurt performance.

Editorial extensions

If this is right

  • Multi-task test-time training should be evaluated with synchronization metrics (step variance, DTW, cosine similarity) in addition to per-task gains, because the paper shows these metrics correlate with adaptation performance.
  • S4T's adaptation branch is separable from the main decoders, so it can be attached to existing multi-task architectures; the paper's comparison with a stronger MTL+TTT baseline uses far fewer parameters than that baseline.
  • Because it couples tasks through learned relations rather than through a single auxiliary loss, S4T keeps improving with longer adaptation while several baselines degrade, which is useful when the number of test-time steps is not known in advance.
  • Adapting all tasks jointly with learned relations yields larger gains than adapting a single task, supporting the paper's claim that the relation among tasks, not merely the multi-head architecture, drives the benefit.

Reading between the lines

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

  • Beyond the paper's benchmarks, one can test whether the TBS's learned relations transfer across source-target pairs and across task sets; if relations are the transferable object, a synchronizer trained on one source should accelerate adaptation on an unrelated source with the same target.
  • The paper's affinity-gap plot in Appendix C is a post hoc check, not a guarantee; a testable extension is to estimate the transformation $f$ explicitly and check whether it generalizes to new domain shifts rather than only to the two dataset pairs tested.
  • The paper's Gaussian-noise ablation shows gains shrink as the noise scale grows, suggesting an extension that identifies where the preservation-of-task-relation assumption breaks and what class of transformations $f$ can express.
  • If relation-based synchronization is as central as the experiments suggest, multi-task test-time methods could be classified by how they model task coupling, and hybrid methods that combine relation loss with entropy minimization may be more robust in single-task settings as well.
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 / 3 minor

Summary. The paper proposes S4T, a test-time training method for multi-task models. It introduces a Task Behavior Synchronizer (TBS) that, during source training, predicts task labels from masked task-specific latent vectors, and during test time minimizes a pseudo-label consistency loss between the TBS output and the main task decoder predictions. The authors argue that predicting task relations across domain shifts is the key to synchronizing multiple tasks during adaptation, and they provide a theoretical proposition intended to show that minimizing this consistency loss reduces the target-domain supervised loss. Empirically, the paper evaluates S4T on Taskonomy-to-NYUD-v2 and Taskonomy-to-PASCAL-Context shifts, with additional reverse-shift experiments in the appendix, and reports consistent improvements over prior TTA/TTT methods, plus synchronization metrics and ablations.

Significance. If the empirical results hold, S4T is a practically useful contribution to multi-task test-time training: it is modular, works on dense prediction tasks, and appears to avoid the performance collapse observed in other TTT methods under longer adaptation. The paper also ships its source code, reports synchronization metrics, and includes ablations that support the component design. However, the theoretical justification in Section 3.2 is not yet sound as stated: Proposition 1 omits a term from the appendix derivation, and the proposed test-time objective is not shown to control the target supervised loss. Since the theoretical explanation is one of the paper's stated contributions, this needs to be fixed before the manuscript can be accepted.

major comments (3)
  1. [Section 3.2 / Appendix E, Eq. (22)] Proposition 1 as printed in Section 3.2 omits the third term d[p(yt,j|{zt,i}), p(yt,j|{~zt,i})] that appears as Eq. (22) in Appendix E. The displayed two-term bound is therefore not the actual consequence of the derivation. The full inequality contains a target-domain relation gap that Assumption 2 only controls on the source domain, so the main text should either state the full bound or provide an explicit argument for why this third term is negligible at test time.
  2. [Section 3.2, Eq. (3) and Section 3.3, Eq. (8)] The test-time objective minimizes only the expectation of D, but the right-hand side of Proposition 1 also contains the masked-target supervised term d(θ, p({~zt,i}, yt,j)), which is not observed and is not minimized. If the model were to collapse so that pθ(y|z) and pθ(y|~z) both become the same constant distribution, then E[D] = 0 while the target supervised loss can be arbitrarily large; the inequality would then be vacuous. The paper needs additional assumptions or a weaker claim to show that minimizing Eq. (3) reduces target task error.
  3. [Appendix E, Eqs. (25)-(28)] The step from Eq. (25) to Eq. (28) is not implied by Assumption 1 as stated. It requires that the target-domain relation gap be bounded by a constant C times the corresponding source-domain gap uniformly across all tasks and all latent/masked-latent pairs, which is a quasi-isometry property of the transformation f plus an alignment condition on latent marginals. Neither condition is formalized in Assumption 1 nor checked experimentally, and the phrase 'with a scaling factor C' introduces a new assumption inside the derivation.
minor comments (3)
  1. [Section 4.2] The protocol 'select the point at which each method achieves its best TTT performance' is not a fixed-budget comparison and can favor methods with a single good early step. The paper should also report results at a common number of adaptation steps for all methods, as Figure 4 suggests such a comparison is feasible.
  2. [Tables 1, 7, and Appendix A] The tables report standard deviations but do not state the number of seeds or independent runs used, which is needed to interpret the significance of the differences.
  3. [Appendix C, Fig. 6] The affinity-gap plot supports Assumption 1 only after adaptation has already taken place; it would be more convincing to measure the relation gap before adaptation and to show that the source relation predicts the target relation without post-hoc adjustment.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the empirical TTT claim is independently benchmarked, and the theory gaps are non-circular.

full rationale

The central empirical claim is self-contained: the Task Behavior Synchronizer is trained with source-domain ground-truth labels, and at test time S4T minimizes a consistency loss between its own masked and unmasked predictions without access to target labels; the reported gains (Table 1) are then evaluated against held-out target metrics (mIoU, RMSE, mErr). Assumptions 1 and 2 are explicit modeling assumptions rather than definitions of the evaluation outcome, and Appendix C checks them post hoc through affinity-gap measurements and mask-ratio robustness, providing evidence external to the fitted objective. The self-citations [16,17] appear only in related-work context and are not load-bearing for the method's design or its claimed results; there is no invoked author-imported uniqueness theorem. The main concern in the derivation chain is that Proposition 1's bound includes an unminimized masked target-supervised term (Eq. (1)/Eq. (20)), and the step from Eq. (25) to Eq. (28) requires an unstated uniform quasi-isometry condition on the transformation f. This is a theoretical-support gap and a soundness risk, but it is not a case of a prediction reducing by construction to a fitted input or to a self-citation, so it does not constitute circularity under the scoring rubric.

Assumptions & free parameters 2 free parameters · 3 assumptions · 0 invented entities

The central method rests on two explicitly stated statistical assumptions about transfer of task relations and adequacy of the TBS, plus an implicit bound on transferring the relation gap. Free parameters are limited to loss weights and the masking ratio, with the masking ratio tuned on target-domain performance. No new physical or ontological entities are introduced; the Task Behavior Synchronizer is a learned module, not a postulated entity.

free parameters (2)
  • masking_ratio = 0.7-0.8
    Selected by peaking the target-domain metric (delta TTT) on Taskonomy to NYUD-v2 in Appendix D, Fig. 8. This is tuning to the evaluation benchmark, not a parameter-free design choice.
  • loss_weights lambda_TBS and lambda_TP = lambda_TBS = 1 during train, 0.01 during test; lambda_TP = 1
    Manually chosen hyperparameters in Appendix A. The balance between the main loss, the TBS loss, and the task-projection loss affects how well task relations are learned and how strongly they drive adaptation.
assumptions (3)
  • domain assumption Assumption 1: structural dependencies between latent variables and task labels remain consistent across domains, p(ys,i|zs,1,...,zs,n) approximately equals f(p(yt,i|zt,1,...,zt,n)).
    Introduced in Section 3.2. The entire transfer mechanism depends on task relations persisting under domain shift; if they do not, the source-trained TBS may mislead target adaptation. The paper gives a post-hoc affinity-gap check in Appendix C but no guarantee.
  • ad hoc to paper Assumption 2: if the TBS is sufficiently trained, p(yi|z1,...,zn) approximately equals p(yi|z~1,...,z~n).
    Introduced in Section 3.2. This assumption is specific to the paper's own module and is used in Appendix E to drop the source-domain relation gap. It effectively requires the masked and unmasked predictions to already agree, which is close to the behavior the method is designed to produce.
  • ad hoc to paper Bounded transfer of the relation gap: the target-domain ideal relation gap is bounded by a constant C times the source-domain relation gap.
    Used in Appendix E, Eq. (25) to Eq. (28), to remove the third term and obtain the clean two-term Proposition 1. The constant C is unspecified and the bound is asserted rather than derived from Assumption 1 as stated.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Synchronizing Task Behavior: Aligning Multiple Tasks during Test-Time Training." pith.science (2026). https://pith.science/paper/CMXOPHXJ

@misc{pith2026250707778,
  author       = {Pith},
  title        = {Pith review of: Synchronizing Task Behavior: Aligning Multiple Tasks during Test-Time Training},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CMXOPHXJ}},
  note         = {Machine review of arXiv:2507.07778}
}
read the original abstract

Generalizing neural networks to unseen target domains is a significant challenge in real-world deployments. Test-time training (TTT) addresses this by using an auxiliary self-supervised task to reduce the domain gap caused by distribution shifts between the source and target. However, we find that when models are required to perform multiple tasks under domain shifts, conventional TTT methods suffer from unsynchronized task behavior, where the adaptation steps needed for optimal performance in one task may not align with the requirements of other tasks. To address this, we propose a novel TTT approach called Synchronizing Tasks for Test-time Training (S4T), which enables the concurrent handling of multiple tasks. The core idea behind S4T is that predicting task relations across domain shifts is key to synchronizing tasks during test time. To validate our approach, we apply S4T to conventional multi-task benchmarks, integrating it with traditional TTT protocols. Our empirical results show that S4T outperforms state-of-the-art TTT methods across various benchmarks.

Figures

Figures reproduced from arXiv: 2507.07778 by the authors.

Figure 1
Figure 1. Comparison of conventional Test-Time Training and the [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Overall framework. The input x is encoded into the latent z by the shared encoder p(z|x). Meanwhile, task-specific projection maps z to a task-specific latent zi. These are stacked and fed into the task-specific decoder and TBS or passed through one more layer to predict another output y T P i . For each task i, the decoder p(yi|zi) generates the main output y main i . The TBS p(yi|{z˜i} n i=1) uses the masked laten… view at source ↗
Figure 3
Figure 3. Comparison of previous TTA and TTT methods with proposed S4T in terms of task synchronization. We evaluate the degree of [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Comparison of previous TTA and TTT methods with our S4T across time steps during test-time training. We evaluate the [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Candidates for masking strategy Mi applied to task￾specific latent vectors, z˜i = Mi(zi). Each task-specific latent vector zi is represented by separate large squares, with the un￾masked portions shaded in black. In (a), randomly select patches for masking. In (b), mas…
Figure 6
Figure 6. Figure 6: Task affinity gap before and after adaptation. [PITH_FULL_IMAGE:figures/full_fig_p013_6.png]
Figure 7
Figure 7. Figure 7: TBS prediction loss across varying masking ratios. [PITH_FULL_IMAGE:figures/full_fig_p013_7.png]
Figure 8
Figure 8. Figure 8: Ablation study on the masking ratio of S4T. We evaluate performance under the domain shift from Taskonomy to NYUD-v2. [PITH_FULL_IMAGE:figures/full_fig_p014_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

64 extracted references · 54 canonical work pages

  1. [1]

    Multitask learning

    Rich Caruana. Multitask learning. Machine learning, 28: 41–75, 1997. 2

    work page 1997

  2. [2]

    Improved test-time adaptation for domain generalization

    Liang Chen, Yong Zhang, Yibing Song, Ying Shan, and Lingqiao Liu. Improved test-time adaptation for domain generalization. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 24172– 24182, 2023. 2

    work page 2023

  3. [3]

    A multi-task mean teacher for semi-supervised shadow detection

    Zhihao Chen, Lei Zhu, Liang Wan, Song Wang, Wei Feng, and Pheng-Ann Heng. A multi-task mean teacher for semi-supervised shadow detection. In Proceedings of the IEEE/CVF Conference on computer vision and pattern recog- nition, pages 5611–5620, 2020. 2

    work page 2020

  4. [4]

    Mod-squad: Designing mixtures of experts as modular multi- task learners

    Zitian Chen, Yikang Shen, Mingyu Ding, Zhenfang Chen, Hengshuang Zhao, Erik G Learned-Miller, and Chuang Gan. Mod-squad: Designing mixtures of experts as modular multi- task learners. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 11828– 11837, 2023. 1, 3, 5

    work page 2023

  5. [5]

    Test-time linear out-of-distribution detection

    Ke Fan, Tong Liu, Xingyu Qiu, Yikai Wang, Lian Huai, Zeyu Shangguan, Shuang Gou, Fengjian Liu, Yuqian Fu, Yanwei Fu, et al. Test-time linear out-of-distribution detection. InPro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 23752–23761, 2024. 2

    work page 2024

  6. [6]

    M3vit: Mixture-of-experts vision transformer for efficient multi-task learning with model-accelerator co-design

    Zhiwen Fan, Rishov Sarkar, Ziyu Jiang, Tianlong Chen, Kai Zou, Yu Cheng, Cong Hao, Zhangyang Wang, et al. M3vit: Mixture-of-experts vision transformer for efficient multi-task learning with model-accelerator co-design. Advances in Neu- ral Information Processing Systems, 35:28441–28457, 2022. 1, 3, 5

    work page 2022

  7. [7]

    Test-time training with masked autoencoders

    Yossi Gandelsman, Yu Sun, Xinlei Chen, and Alexei Efros. Test-time training with masked autoencoders. Advances in Neural Information Processing Systems , 35:29374–29385,

    work page

  8. [8]

    Clust3: Information invariant test-time training

    Gustavo A Vargas Hakim, David Osowiechi, Mehrdad Noori, Milad Cheraghalikhani, Ali Bahri, Ismail Ben Ayed, and Christian Desrosiers. Clust3: Information invariant test-time training. In Proceedings of the IEEE/CVF International Con- ference on Computer Vision, pages 6136–6145, 2023. 2, 5, 6, 14

    work page 2023

Show all 64 references
  1. [9]

    Deep residual learning for image recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In Proceed- ings of the IEEE conference on computer vision and pattern recognition, pages 770–778, 2016. 12

    2016

  2. [10]

    Masked autoencoders are scalable vision learners

    Kaiming He, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Doll´ar, and Ross Girshick. Masked autoencoders are scalable vision learners. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 16000– 16009, 2022. 1, 2, 3

    2022

  3. [11]

    Denoising diffu- sion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffu- sion probabilistic models. Advances in neural information processing systems, 33:6840–6851, 2020. 3

    2020

  4. [12]

    Partly supervised multi-task learning

    Chao Huang, Hui Tang, Wei Fan, Yuan Xiao, Dingjun Hao, Zhen Qian, Demetri Terzopoulos, et al. Partly supervised multi-task learning. In 2020 19th IEEE International Confer- ence on Machine Learning and Applications (ICMLA), pages 769–774. IEEE, 2020. 2

    2020

  5. [13]

    Going beyond multi-task dense prediction with synergy em- bedding models

    Huimin Huang, Yawen Huang, Lanfen Lin, Ruofeng Tong, Yen-Wei Chen, Hao Zheng, Yuexiang Li, and Yefeng Zheng. Going beyond multi-task dense prediction with synergy em- bedding models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 28...

    2024

  6. [14]

    Semi- supervised multi-task learning with chest x-ray images

    Abdullah-Al-Zubaer Imran and Demetri Terzopoulos. Semi- supervised multi-task learning with chest x-ray images. In Machine Learning in Medical Imaging: 10th International Workshop, MLMI 2019, Held in Conjunction with MICCAI 2019, Shenzhen, China, October 13, 2019, Proceedings ...

    2019

  7. [15]

    Test-time classifier adjustment module for model-agnostic domain generalization

    Yusuke Iwasawa and Yutaka Matsuo. Test-time classifier adjustment module for model-agnostic domain generalization. Advances in Neural Information Processing Systems, 34:2427– 2440, 2021. 2

    2021

  8. [16]

    Quantifying task priority for multi-task optimization

    Wooseong Jeong and Kuk-Jin Yoon. Quantifying task priority for multi-task optimization. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 363–372, 2024. 2

    2024

  9. [17]

    Selective task group updates for multi-task optimization

    Wooseong Jeong and Kuk-Jin Yoon. Selective task group updates for multi-task optimization. arXiv preprint arXiv:2502.11986, 2025

    2025 arXiv

  10. [18]

    Multi-task learning using uncertainty to weigh losses for scene geometry and semantics

    Alex Kendall, Yarin Gal, and Roberto Cipolla. Multi-task learning using uncertainty to weigh losses for scene geometry and semantics. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 7482–7491,

  11. [19]

    Adam: A method for stochastic optimization

    Diederik P Kingma and Jimmy Ba. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 ,

    arXiv

  12. [20]

    Multi-task semi-supervised ad- versarial autoencoding for speech emotion recognition

    Siddique Latif, Rajib Rana, Sara Khalifa, Raja Jurdak, Julien Epps, and Bj¨orn W Schuller. Multi-task semi-supervised ad- versarial autoencoding for speech emotion recognition. IEEE Transactions on Affective computing, 13(2):992–1004, 2020. 2

    2020

  13. [21]

    Becotta: Input-dependent online blending of experts for continual test- time adaptation

    Daeun Lee, Jaehong Yoon, and Sung Ju Hwang. Becotta: Input-dependent online blending of experts for continual test- time adaptation. arXiv preprint arXiv:2402.08712, 2024. 2

    2024 arXiv

  14. [22]

    Learning multiple dense prediction tasks from partially annotated data

    Wei-Hong Li, Xialei Liu, and Hakan Bilen. Learning multiple dense prediction tasks from partially annotated data. In Pro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 18879–18889, 2022. 2

    2022

  15. [23]

    On the robustness of open-world test-time training: Self-training with dynamic prototype expansion

    Yushu Li, Xun Xu, Yongyi Su, and Kui Jia. On the robustness of open-world test-time training: Self-training with dynamic prototype expansion. In Proceedings of the IEEE/CVF In- ternational Conference on Computer Vision, pages 11836– 11846, 2023. 2

    2023

  16. [24]

    A comprehensive survey on test-time adaptation under distribution shifts.International Journal of Computer Vision, pages 1–34, 2024

    Jian Liang, Ran He, and Tieniu Tan. A comprehensive survey on test-time adaptation under distribution shifts.International Journal of Computer Vision, pages 1–34, 2024. 2

    2024

  17. [25]

    Auxiliary tasks in multi-task learning

    Lukas Liebel and Marco K¨orner. Auxiliary tasks in multi-task learning. arXiv preprint arXiv:1805.06334, 2018. 1

    2018 arXiv

  18. [26]

    Conflict-averse gradient descent for multi-task learn- ing

    Bo Liu, Xingchao Liu, Xiaojie Jin, Peter Stone, and Qiang Liu. Conflict-averse gradient descent for multi-task learn- ing. Advances in Neural Information Processing Systems, 34: 18878–18890, 2021. 2 9

    2021

  19. [27]

    Famo: Fast adaptive multitask optimization

    Bo Liu, Yihao Feng, Peter Stone, and Qiang Liu. Famo: Fast adaptive multitask optimization. Advances in Neural Information Processing Systems, 36, 2024. 2

    2024

  20. [28]

    Semi- supervised multitask learning

    Qiuhua Liu, Xuejun Liao, and Lawrence Carin. Semi- supervised multitask learning. Advances in Neural Infor- mation Processing Systems, 20, 2007. 2

    2007

  21. [29]

    Ttt++: When does self-supervised test-time training fail or thrive? Advances in Neural Information Processing Systems , 34: 21808–21820, 2021

    Yuejiang Liu, Parth Kothari, Bastien Van Delft, Baptiste Bellot-Gurlet, Taylor Mordan, and Alexandre Alahi. Ttt++: When does self-supervised test-time training fail or thrive? Advances in Neural Information Processing Systems , 34: 21808–21820, 2021. 1, 2, 5, 6, 14

    2021

  22. [30]

    Taskology: Utilizing task relations at scale

    Yao Lu, Soren Pirk, Jan Dlabal, Anthony Brohan, Ankita Pasad, Zhao Chen, Vincent Casser, Anelia Angelova, and Ariel Gordon. Taskology: Utilizing task relations at scale. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 8700–8709, 2021. 2

    2021

  23. [31]

    Attentive single-tasking of multiple tasks

    Kevis-Kokitsi Maninis, Ilija Radosavovic, and Iasonas Kokki- nos. Attentive single-tasking of multiple tasks. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 1851–1860, 2019. 1, 5

    2019

  24. [32]

    Act- mad: Activation matching to align distributions for test- time-training

    Muhammad Jehanzeb Mirza, Pol Jan´e Soneira, Wei Lin, Ma- teusz Kozinski, Horst Possegger, and Horst Bischof. Act- mad: Activation matching to align distributions for test- time-training. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, page...

    2023

  25. [33]

    The role of context for object detection and semantic segmentation in the wild

    Roozbeh Mottaghi, Xianjie Chen, Xiaobai Liu, Nam-Gyu Cho, Seong-Whan Lee, Sanja Fidler, Raquel Urtasun, and Alan Yuille. The role of context for object detection and semantic segmentation in the wild. In Proceedings of the IEEE conference on computer vision and pattern recogni...

    2014

  26. [34]

    Multimodal contrastive learning with limoe: the language-image mixture of experts

    Basil Mustafa, Carlos Riquelme, Joan Puigcerver, Rodolphe Jenatton, and Neil Houlsby. Multimodal contrastive learning with limoe: the language-image mixture of experts. Advances in Neural Information Processing Systems , 35:9564–9576,

  27. [35]

    Multi- task learning as a bargaining game

    Aviv Navon, Aviv Shamsian, Idan Achituve, Haggai Maron, Kenji Kawaguchi, Gal Chechik, and Ethan Fetaya. Multi- task learning as a bargaining game. arXiv preprint arXiv:2202.01017, 2022. 2

    2022 arXiv

  28. [36]

    Tipi: Test time adaptation with transformation invariance

    A Tuan Nguyen, Thanh Nguyen-Tang, Ser-Nam Lim, and Philip HS Torr. Tipi: Test time adaptation with transformation invariance. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 24162– 24171, 2023. 1, 2, 5, 6, 14

    2023

  29. [37]

    Joint-task regularization for partially labeled multi-task learn- ing

    Kento Nishi, Junsik Kim, Wanhua Li, and Hanspeter Pfister. Joint-task regularization for partially labeled multi-task learn- ing. In Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition, pages 16152–16162,

  30. [38]

    Tttflow: Unsupervised test-time training with normalizing flow

    David Osowiechi, Gustavo A Vargas Hakim, Mehrdad Noori, Milad Cheraghalikhani, Ismail Ben Ayed, and Christian Desrosiers. Tttflow: Unsupervised test-time training with normalizing flow. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pages ...

    2023

  31. [39]

    Nc-ttt: A noise constrastive approach for test-time training

    David Osowiechi, Gustavo A Vargas Hakim, Mehrdad Noori, Milad Cheraghalikhani, Ali Bahri, Moslem Yazdanpanah, Ismail Ben Ayed, and Christian Desrosiers. Nc-ttt: A noise constrastive approach for test-time training. In Proceedings of the IEEE/CVF Conference on Computer Vision a...

    2024

  32. [40]

    Variational inference with normalizing flows

    Danilo Rezende and Shakir Mohamed. Variational inference with normalizing flows. In International conference on ma- chine learning, pages 1530–1538. PMLR, 2015. 2

    2015

  33. [41]

    Scaling vision with sparse mix- ture of experts

    Carlos Riquelme, Joan Puigcerver, Basil Mustafa, Maxim Neumann, Rodolphe Jenatton, Andr´e Susano Pinto, Daniel Keysers, and Neil Houlsby. Scaling vision with sparse mix- ture of experts. Advances in Neural Information Processing Systems, 34:8583–8595, 2021. 3, 5

    2021

  34. [42]

    Learning to relate depth and semantics for unsuper- vised domain adaptation

    Suman Saha, Anton Obukhov, Danda Pani Paudel, Menelaos Kanakis, Yuhua Chen, Stamatios Georgoulis, and Luc Van Gool. Learning to relate depth and semantics for unsuper- vised domain adaptation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition...

    2021

  35. [43]

    Independent component alignment for multi- task learning

    Dmitry Senushkin, Nikolay Patakin, Arseny Kuznetsov, and Anton Konushin. Independent component alignment for multi- task learning. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 20083– 20093, 2023. 2

    2023

  36. [44]

    Indoor segmentation and support inference from rgbd images

    Nathan Silberman, Derek Hoiem, Pushmeet Kohli, and Rob Fergus. Indoor segmentation and support inference from rgbd images. In Computer Vision–ECCV 2012: 12th European Conference on Computer Vision, Florence, Italy, October 7- 13, 2012, Proceedings, Part V 12, pages 746–760. Springer,

    2012

  37. [45]

    Revisiting realistic test-time training: Sequential inference and adaptation by anchored clustering

    Yongyi Su, Xun Xu, and Kui Jia. Revisiting realistic test-time training: Sequential inference and adaptation by anchored clustering. Advances in Neural Information Processing Sys- tems, 35:17543–17555, 2022. 2

    2022

  38. [46]

    Test-time training with self- supervision for generalization under distribution shifts

    Yu Sun, Xiaolong Wang, Zhuang Liu, John Miller, Alexei Efros, and Moritz Hardt. Test-time training with self- supervision for generalization under distribution shifts. In International conference on machine learning, pages 9229–

  39. [47]

    Mti-net: Multi-scale task interaction networks for multi-task learning

    Simon Vandenhende, Stamatios Georgoulis, and Luc Van Gool. Mti-net: Multi-scale task interaction networks for multi-task learning. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part IV 16, pages 527–543. Springer, 2020. 1, 6, 13

    2020

  40. [48]

    On the road to online adaptation for semantic image segmentation

    Riccardo V olpi, Pau De Jorge, Diane Larlus, and Gabriela Csurka. On the road to online adaptation for semantic image segmentation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 19184–19195,

  41. [49]

    Tent: Fully test-time adaptation by entropy minimization

    Dequan Wang, Evan Shelhamer, Shaoteng Liu, Bruno Ol- shausen, and Trevor Darrell. Tent: Fully test-time adaptation by entropy minimization. arXiv preprint arXiv:2006.10726,

    2006 arXiv

  42. [50]

    Semi-supervised multi- task learning with task regularizations

    Fei Wang, Xin Wang, and Tao Li. Semi-supervised multi- task learning with task regularizations. In 2009 Ninth IEEE 10 International Conference on Data Mining , pages 562–568. IEEE, 2009. 2

    2009

  43. [51]

    Continual test-time domain adaptation

    Qin Wang, Olga Fink, Luc Van Gool, and Dengxin Dai. Continual test-time domain adaptation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 7201–7211, 2022. 2

    2022

  44. [52]

    Semi-supervised multi- task learning for semantics and depth

    Yufeng Wang, Yi-Hsuan Tsai, Wei-Chih Hung, Wenrui Ding, Shuo Liu, and Ming-Hsuan Yang. Semi-supervised multi- task learning for semantics and depth. In Proceedings of the IEEE/CVF winter conference on applications of computer vision, pages 2505–2514, 2022. 2

    2022

  45. [53]

    Multi-task dense prediction via mixture of low-rank experts

    Yuqi Yang, Peng-Tao Jiang, Qibin Hou, Hao Zhang, Jinwei Chen, and Bo Li. Multi-task dense prediction via mixture of low-rank experts. In Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pages 27927–27937, 2024. 1, 6

    2024

  46. [54]

    Inverted pyramid multi-task trans- former for dense scene understanding

    Hanrong Ye and Dan Xu. Inverted pyramid multi-task trans- former for dense scene understanding. In Computer Vision– ECCV 2022: 17th European Conference, Tel Aviv, Israel, October 23–27, 2022, Proceedings, Part XXVII, pages 514–

    2022

  47. [55]

    Taskprompter: Spatial-channel multi-task prompting for dense scene understanding

    Hanrong Ye and Dan Xu. Taskprompter: Spatial-channel multi-task prompting for dense scene understanding. In The Eleventh International Conference on Learning Representa- tions, 2022. 1, 6

    2022

  48. [56]

    Diffusionmtl: Learning multi- task denoising diffusion model from partially annotated data

    Hanrong Ye and Dan Xu. Diffusionmtl: Learning multi- task denoising diffusion model from partially annotated data. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 27960–27969, 2024. 3

    2024

  49. [57]

    Gradient surgery for multi- task learning

    Tianhe Yu, Saurabh Kumar, Abhishek Gupta, Sergey Levine, Karol Hausman, and Chelsea Finn. Gradient surgery for multi- task learning. Advances in neural information processing systems, 33:5824–5836, 2020. 2

    2020

  50. [58]

    Taskonomy: Disentangling task transfer learning

    Amir R Zamir, Alexander Sax, William Shen, Leonidas J Guibas, Jitendra Malik, and Silvio Savarese. Taskonomy: Disentangling task transfer learning. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 3712–3722, 2018. 5

    2018

  51. [59]

    Robust learning through cross-task consistency

    Amir R Zamir, Alexander Sax, Nikhil Cheerla, Rohan Suri, Zhangjie Cao, Jitendra Malik, and Leonidas J Guibas. Robust learning through cross-task consistency. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 11197–11206, 2020. 2

    2020

  52. [60]

    Generaliz- able model-agnostic semantic segmentation via target-specific normalization

    Jian Zhang, Lei Qi, Yinghuan Shi, and Yang Gao. Generaliz- able model-agnostic semantic segmentation via target-specific normalization. Pattern Recognition, 122:108292, 2022. 2

    2022

  53. [61]

    Mixture of attention heads: Selecting attention heads per token

    Xiaofeng Zhang, Yikang Shen, Zeyu Huang, Jie Zhou, Wenge Rong, and Zhang Xiong. Mixture of attention heads: Selecting attention heads per token. arXiv preprint arXiv:2210.05144, 2022. 3, 5

    2022 arXiv

  54. [62]

    Semi-supervised multi-task regression

    Yu Zhang and Dit-Yan Yeung. Semi-supervised multi-task regression. In Machine Learning and Knowledge Discovery in Databases: European Conference, ECML PKDD 2009, Bled, Slovenia, September 7-11, 2009, Proceedings, Part II 20, pages 617–631. Springer, 2009. 2

    2009

  55. [63]

    Pattern-affinitive propagation across depth, surface normal and semantic segmentation

    Zhenyu Zhang, Zhen Cui, Chunyan Xu, Yan Yan, Nicu Sebe, and Jian Yang. Pattern-affinitive propagation across depth, surface normal and semantic segmentation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 4106–4115, 2019. 1, 6 11 Sy...

    2019

  56. [9248]

    1, 2, 5, 6, 13, 14

    PMLR, 2020. 1, 2, 5, 6, 13, 14

    2020

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.