mlsanityData to 14 Oct 2026

TecoPrompt: Temporal-Conservative Prompt Learning for Vision-Language Models

arXiv 2609.16858 ↗First posted 15 Sep 2026cs.CVcited by 0

On the map: Multimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP

Abstract (arXiv)

Prompt learning adapts vision-language models, such as CLIP, by adjusting a small set of context tokens. However, under few-shot supervision, even moderate label noise can disrupt prompt optimization. To address this issue, we propose TecoPrompt, a closed-loop robust prompt-learning framework that revisits optimal transport (OT) pseudo-labeling from a temporal perspective. TecoPrompt employs an entropic OT plan in the CLIP semantic space to obtain globally consistent label candidates. It verifies the reliability of these candidates by examining trajectory stability: a noisy label is only rewritten if the OT candidate remains unchanged within a K-epoch temporal stability window and passes a confidence gate based on Exponential Moving Average (EMA). This approach helps reduce confirmation bias. The rewritten labels are then integrated back into prompt training using a tri-group objective that includes three loss functions aligned with clean, mid, and noisy subsets. Experiments on seven datasets with synthetic symmetric and asymmetric noise, as well as Food101N, demonstrate significant performance improvements. For example, on the OxfordPets dataset, with 50% asymmetric noise, TecoPrompt achieves an accuracy of 0.843, up from 0.775.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Prompt learning and adaptation of CLIP, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 41 of its 52 references that are in the map. First posted on arXiv 15 Sep 2026.

Builds on

41 of 52 references.

Closest ancestors · 9

Same direction, newest first

  1. 2024NLPrompt: Noise-Label Prompt Learning for Vision-Language Models2 Dec
  2. 2024GalLoP: Learning Global and Local Prompts for Vision-Language Models1 Jul
  3. 2024C-TPT: Calibrated Test-Time Prompt Tuning for Vision-Language Models via Text Feature Dispersion21 Mar
  4. 2024PromptKD: Unsupervised Prompt Distillation for Vision-Language Models5 Mar
  5. 2023Why Is Prompt Tuning for Vision-Language Models Robust to Noisy Labels?22 Jul
  6. 2022Test-Time Prompt Tuning for Zero-Shot Generalization in Vision-Language Models15 Sep
  7. 2022Prompt Distribution Learning6 May
  8. 2022Conditional Prompt Learning for Vision-Language Models10 Mar
  9. 2021Learning to Prompt for Vision-Language Models2 Sep

Other references · 32

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  3. 2015Distilling the Knowledge in a Neural Networkcited by 10,324
  4. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,475
  5. 2021Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervisioncited by 3,864
  6. 2013Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distancescited by 2,316
  7. 2013Describing Textures in the Wildcited by 2,125
  8. 2022Visual Prompt Tuningcited by 1,681
  9. 2016Understanding deep learning requires rethinking generalizationcited by 1,674
  10. 2017EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classificationcited by 1,496
  11. 2022CoCa: Contrastive Captioners are Image-Text Foundation Modelscited by 1,274
  12. 2018Co-teaching: Robust Training of Deep Neural Networks with Extremely Noisy Labelscited by 1,059
20 more
  1. 2016Temporal Ensembling for Semi-Supervised Learningcited by 1,057
  2. 2018Generalized Cross Entropy Loss for Training Deep Neural Networks with Noisy Labelscited by 997
  3. 2017A Closer Look at Memorization in Deep Networkscited by 942
  4. 2018Computational Optimal Transportcited by 780
  5. 2020DivideMix: Learning with Noisy Labels as Semi-supervised Learningcited by 692
  6. 2022MaPLe: Multi-modal Prompt Learningcited by 687
  7. 2018An Empirical Study of Example Forgetting during Deep Neural Network Learningcited by 678
  8. 2016Making Deep Neural Networks Robust to Label Noise: a Loss Correction Approachcited by 665
  9. 2019Using Pre-Training Can Improve Model Robustness and Uncertaintycited by 373
  10. 2014Training Deep Neural Networks on Noisy Labels with Bootstrappingcited by 322
  11. 2020Combating noisy labels by agreement: A joint training method with co-regularizationcited by 289
  12. 2023Self-regulating Prompts: Foundational Model Adaptation without Forgettingcited by 272
  13. 2017CleanNet: Transfer Learning for Scalable Image Classifier Training with Label Noisecited by 181
  14. 2020Dual T: Reducing Estimation Error for Transition Matrix in Label-noise Learningcited by 142
  15. 2019Curriculum Loss: Robust Learning and Generalization against Label Corruptioncited by 88
  16. 2021Learning Noise Transition Matrix from Only Noisy Labels via Total Variation Regularizationcited by 62
  17. 2022Few-shot Learning with Noisy Labelscited by 23
  18. 2021Two Wrongs Don't Make a Right: Combating Confirmation Bias in Learning with Label Noisecited by 18
  19. 2023CSOT: Curriculum and Structure-Aware Optimal Transport for Learning with Noisy Labelscited by 17
  20. 2018When Will Gradient Methods Converge to Max-margin Classifier under ReLU Models?cited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Seeing is Believing: Robust Vision-Guided Cross-Modal Prompt Learning under Label Noise
  2. 2024Tuning Vision-Language Models with Candidate Labels by Prompt Alignment
  3. 2025Prompt-OT: An Optimal Transport Regularization Paradigm for Knowledge Preservation in Vision-Language Model Adaptation
  4. 2025Robust Prompt Tuning for Vision-Language Models with Mild Semantic Noise
  5. 2023Patch-Prompt Aligned Bayesian Prompt Tuning for Vision-Language Models
  6. 2025Alternating Training-based Label Smoothing Enhances Prompt Generalization
  7. 2026Noise-aware few-shot learning through bi-directional multi-view prompt alignment
  8. 2025Prompting without Panic: Attribute-aware, Zero-shot, Test-Time Calibration