paper-with-me

홈 › Papers

D-TPT: Dimensional Entropy Maximization for Calibrating Test-Time Prompt Tuning in Vision-Language Models

2025-10-10 · Jisu Han, Wonjun Hwang arxiv

Test-time adaptation paradigm provides flexibility towards domain shifts by performing immediate adaptation on unlabeled target data from the source model. Vision-Language Models (VLMs) leverage their generalization capabilities for diverse downstream tasks, and test-time prompt tuning has emerged as a prominent solution for adapting VLMs. In this work, we explore contrastive VLMs and identify the modality gap caused by a single dominant feature dimension across modalities. We observe that the dominant dimensions in both text and image modalities exhibit high predictive sensitivity, and that constraining their influence can improve calibration error. Building on this insight, we propose dimensional entropy maximization that regularizes the distribution of textual features toward uniformity to mitigate the dependency of dominant dimensions. Our method alleviates the degradation of calibration performance in test-time prompt tuning, offering a simple yet effective solution to enhance the reliability of VLMs in real-world deployment scenarios.

📄 PDF Abstract BibTeX arXiv:2510.09473

Code (0)

등록된 구현이 없습니다.

Tasks

Test-time Adaptation

Similar Papers 제목 키워드 기반

Test-Time Adaptation to Distribution Shift by Confidence Maximization and Input Transformation

2021-06-28 · Chaithanya Kumar Mummadi, Robin Hutmacher, Kilian Rambach, Evgeny Levinkov 외

Deep neural networks often exhibit poor performance on data that is unlikely under the train-time data distribution, for instance data affected by corruptions. Previous works demonstrate that test-time adaptation to data…

Test-time Adaptation

Test-Time Adaptation to Distribution Shifts by Confidence Maximization and Input Transformation

2021-05-21 · NeurIPS 2021 12 · Chaithanya Kumar Mummadi, Robin Hutmacher, Kilian Rambach, Evgeny Levinkov 외

Deep neural networks often exhibit poor performance on data that is unlikely under the train-time data distribution, for instance data affected by corruptions. Previous works demonstrate that test-time adaptation to data…

Test-time Adaptation

Improving Pre-trained Self-Supervised Embeddings Through Effective Entropy Maximization

2024-11-24 · Deep Chakraborty, Yann Lecun, Tim G. J. Rudner, Erik Learned-Miller

A number of different architectures and loss functions have been applied to the problem of self-supervised learning (SSL), with the goal of developing embeddings that provide the best possible pre-training for as-yet-unk…

Self-Supervised Learning

Mitigating the ID-OOD Tradeoff in Open-Set Test-Time Adaptation

2026-04-02 · Wenjie Zhao, Jia Li, Xin Dong, Yapeng Tian 외 arxiv

Open-set test-time adaptation (OSTTA) addresses the challenge of adapting models to new environments where out-of-distribution (OOD) samples coexist with in-distribution (ID) samples affected by distribution shifts. In s…

Semantic SegmentationTest-time Adaptation

Unsupervised Domain Adaptation via Calibrating Uncertainties

2019-07-25 · Ligong Han, Yang Zou, Ruijiang Gao, Lezi Wang 외

Unsupervised domain adaptation (UDA) aims at inferring class labels for unlabeled target domain given a related labeled source dataset. Intuitively, a model trained on source domain normally produces higher uncertainties…

Domain AdaptationUnsupervised Domain Adaptation