# Test-Time Adaptation via Cache Personalization for Facial Expression Recognition in Videos

arXiv: https://arxiv.org/abs/2603.21309 · HTML: https://mlsanity.com/paper/2603.21309/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 44 of its 62 references that are in the map.
- First posted on arXiv 22 Mar 2026.

## Builds on (most cited first, 30 of 44)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models](https://mlsanity.com/paper/2301.12597/) (2023)
- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [CoCa: Contrastive Captioners are Image-Text Foundation Models](https://mlsanity.com/paper/2205.01917/) (2022)
- [CLIP-Adapter: Better Vision-Language Models with Feature Adapters](https://mlsanity.com/paper/2110.04544/) (2021)
- [Continual Test-Time Domain Adaptation](https://mlsanity.com/paper/2203.13591/) (2022)
- [Efficient Test-Time Model Adaptation without Forgetting](https://mlsanity.com/paper/2204.02610/) (2022)
- [Towards Stable Test-Time Adaptation in Dynamic Wild World](https://mlsanity.com/paper/2302.12400/) (2023)
- [Supervision Exists Everywhere: A Data Efficient Contrastive Language-Image Pre-training Paradigm](https://mlsanity.com/paper/2110.05208/) (2021)
- [Test-Time Prompt Tuning for Zero-Shot Generalization in Vision-Language Models](https://mlsanity.com/paper/2209.07511/) (2022)
- [MEMO: Test Time Robustness via Adaptation and Augmentation](https://mlsanity.com/paper/2110.09506/) (2021)
- [Tip-Adapter: Training-free Adaption of CLIP for Few-shot Classification](https://mlsanity.com/paper/2207.09519/) (2022)
- [Expanding Language-Image Pretrained Models for General Video Recognition](https://mlsanity.com/paper/2208.02816/) (2022)
- [Robust Test-Time Adaptation in Dynamic Scenarios](https://mlsanity.com/paper/2303.13899/) (2023)
- [Diverse Data Augmentation with Diffusions for Effective Test-time Prompt Tuning](https://mlsanity.com/paper/2308.06038/) (2023)
- [Efficient Test-Time Adaptation of Vision-Language Models](https://mlsanity.com/paper/2403.18293/) (2024)
- [Source-free Domain Adaptation via Avatar Prototype Generation and Adaptation](https://mlsanity.com/paper/2106.15326/) (2021)
- [Align Your Prompts: Test-Time Prompting with Distribution Alignment for Zero-Shot Generalization](https://mlsanity.com/paper/2311.01459/) (2023)
- [CLIPood: Generalizing CLIP to Out-of-Distributions](https://mlsanity.com/paper/2302.00864/) (2023)
- [C-TPT: Calibrated Test-Time Prompt Tuning for Vision-Language Models via Text Feature Dispersion](https://mlsanity.com/paper/2403.14119/) (2024)
- [VDM-DA: Virtual Domain Modeling for Source Data-free Domain Adaptation](https://mlsanity.com/paper/2103.14357/) (2021)
- [Prompting Visual-Language Models for Dynamic Facial Expression Recognition](https://mlsanity.com/paper/2308.13382/) (2023)
- [A Closer Look at the Robustness of Contrastive Language-Image Pre-Training (CLIP)](https://mlsanity.com/paper/2402.07410/) (2024)
- [Dual Prototype Evolving for Test-Time Generalization of Vision-Language Models](https://mlsanity.com/paper/2410.12790/) (2024)
- [WATT: Weight Average Test-Time Adaptation of CLIP](https://mlsanity.com/paper/2406.13875/) (2024)
- [EmoCLIP: A Vision-Language Method for Zero-Shot Video Facial Expression Recognition](https://mlsanity.com/paper/2310.16640/) (2023)
- [CLIPER: A Unified Vision-Language Framework for In-the-Wild Facial Expression Recognition](https://mlsanity.com/paper/2303.00193/) (2023)
- [Test-Time Zero-Shot Temporal Action Localization](https://mlsanity.com/paper/2404.05426/) (2024)
- [Leveraging Cross-Modal Neighbor Representation for Improved CLIP Classification](https://mlsanity.com/paper/2404.17753/) (2024)
- [Enhancing Zero-Shot Facial Expression Recognition by LLM Knowledge Transfer](https://mlsanity.com/paper/2405.19100/) (2024)

## Built on by (most cited first, 5 of 5)

- [Multimodal Ambivalence/Hesitancy Recognition in Videos for Personalized Digital Health Interventions](https://mlsanity.com/paper/2604.11730/) (2026)
- [HEDGE: A Calibrated Ensemble for A/H Recognition](https://mlsanity.com/paper/2607.12176/) (2026)
- [CF-Net: Conflict Fusion with Speaker Normalisation and Certainty Weighting for Ambivalence/Hesitancy Recognition](https://mlsanity.com/paper/2607.13976/) (2026)
- [Knowledge-Guided Multimodal Reasoning over Interacting Streams for Video-Level Ambivalence and Hesitancy Recognition](https://mlsanity.com/paper/2607.25961/) (2026)
- [Test-Time Adaptation with Online Personalized Energy-Based Cache for Fine-Grained Video Expression Recognition](https://mlsanity.com/paper/2608.06467/) (2026)

## Abstract (arXiv)

Facial expression recognition (FER) in videos requires model personalization to capture considerable variation across subjects. Vision-language models (VLMs) offer strong transfer through image-text alignment, but their performance can degrade under inter-subject distribution shifts. Test-time adaptation (TTA) can mitigate this challenge, yet most state-of-the-art methods rely on unsupervised parameter optimization, introducing computational overhead that limits real-world deployment. This paper introduces TTA through Cache Personalization (TTA-CaP), a gradient-free, cache-based method for cost-effective personalization of VLMs in video FER. Unlike prior cache-based TTA methods that rely solely on dynamic memories of test samples and may drift because of noisy pseudo-labels, TTA-CaP employs three complementary caches: a personalized static cache constructed through feature-statistics matching, a positive target cache containing reliable subject-specific samples, and a negative target cache containing low-confidence cases as negative evidence. A tri-gate mechanism prevents cache corruption by controlling updates according to temporal stability, confidence, and consistency with the personalized static cache. The caches jointly provide subject-matched positive and negative evidence for robust personalization. TTA-CaP further refines predictions through embedding fusion, supporting temporally stable video-level predictions. Experiments on BioVid, StressID, and BAH show that TTA-CaP outperforms state-of-the-art TTA methods under subject-specific and environmental shifts while maintaining low computational and memory overhead. Our code is publicly available at https://github.com/MasoumehSharafi/TTA-CaP.