Test-Time Adaptation for Tactile-Vision-Language Models
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910025415393280 |
|---|---|
| author | Ye, Chuyang Jing, Haoxian Jiang, Qinting Lin, Yixi Li, Qiang Tang, Xing Jiang, Jingyan |
| author_facet | Ye, Chuyang Jing, Haoxian Jiang, Qinting Lin, Yixi Li, Qiang Tang, Xing Jiang, Jingyan |
| contents | Tactile-vision-language (TVL) models are increasingly deployed in real-world robotic and multimodal perception tasks, where test-time distribution shifts are unavoidable. Existing test-time adaptation (TTA) methods provide filtering in unimodal settings but lack explicit treatment of modality-wise reliability under asynchronous cross-modal shifts, leaving them brittle when some modalities become unreliable. We study TTA for TVL models under such shifts and propose a reliability-aware framework that estimates per-modality reliability from prediction uncertainty and perturbation-based responses. This shared reliability signal is used to (i) filter unreliable test samples, (ii) adaptively fuse tactile, visual, and language features, and (iii) regularize test-time optimization with a reliability-guided objective. On the TAG-C benchmark and additional TVL scenarios, our approach consistently outperforms strong TTA baselines, achieving accuracy gains of up to 49.9\% under severe modality corruptions, underscoring the importance of explicit modality-wise reliability modeling for robust test-time adaptation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_15873 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Test-Time Adaptation for Tactile-Vision-Language Models Ye, Chuyang Jing, Haoxian Jiang, Qinting Lin, Yixi Li, Qiang Tang, Xing Jiang, Jingyan Robotics Artificial Intelligence Tactile-vision-language (TVL) models are increasingly deployed in real-world robotic and multimodal perception tasks, where test-time distribution shifts are unavoidable. Existing test-time adaptation (TTA) methods provide filtering in unimodal settings but lack explicit treatment of modality-wise reliability under asynchronous cross-modal shifts, leaving them brittle when some modalities become unreliable. We study TTA for TVL models under such shifts and propose a reliability-aware framework that estimates per-modality reliability from prediction uncertainty and perturbation-based responses. This shared reliability signal is used to (i) filter unreliable test samples, (ii) adaptively fuse tactile, visual, and language features, and (iii) regularize test-time optimization with a reliability-guided objective. On the TAG-C benchmark and additional TVL scenarios, our approach consistently outperforms strong TTA baselines, achieving accuracy gains of up to 49.9\% under severe modality corruptions, underscoring the importance of explicit modality-wise reliability modeling for robust test-time adaptation. |
| title | Test-Time Adaptation for Tactile-Vision-Language Models |
| topic | Robotics Artificial Intelligence |
| url | https://arxiv.org/abs/2602.15873 |