# Bidirectional Prototype-Reward co-Evolution for Test-Time Adaptation of Vision-Language Models

arXiv: https://arxiv.org/abs/2503.09394 · HTML: https://mlsanity.com/paper/2503.09394/

## Key facts

- Cited by 6 papers in the map (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 36 of its 53 references that are in the map.
- First posted on arXiv 12 Mar 2025.

## Placed in

- Multimodal LLMs / vision-language models / [Prompt learning and adaptation of CLIP](https://mlsanity.com/mllm/prompt-learning-and-adaptation-of-clip/)

## Builds on (most cited first, 30 of 36)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Sigmoid Loss for Language Image Pre-Training](https://mlsanity.com/paper/2303.15343/) (2023)
- [Learning to Prompt for Vision-Language Models](https://mlsanity.com/paper/2109.01134/) (2021)
- [Describing Textures in the Wild](https://mlsanity.com/paper/1311.3618/) (2013)
- [The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalization](https://mlsanity.com/paper/2006.16241/) (2020)
- [Fine-Grained Visual Classification of Aircraft](https://mlsanity.com/paper/1306.5151/) (2013)
- [Do ImageNet Classifiers Generalize to ImageNet?](https://mlsanity.com/paper/1902.10811/) (2019)
- [EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classification](https://mlsanity.com/paper/1709.00029/) (2017)
- [Conditional Prompt Learning for Vision-Language Models](https://mlsanity.com/paper/2203.05557/) (2022)
- [Natural Adversarial Examples](https://mlsanity.com/paper/1907.07174/) (2019)
- [CLIP-Adapter: Better Vision-Language Models with Feature Adapters](https://mlsanity.com/paper/2110.04544/) (2021)
- [Learning Robust Global Representations by Penalizing Local Predictive Power](https://mlsanity.com/paper/1905.13549/) (2019)
- [Vision-Language Models for Vision Tasks: A Survey](https://mlsanity.com/paper/2304.00685/) (2023)
- [MaPLe: Multi-modal Prompt Learning](https://mlsanity.com/paper/2210.03117/) (2022)
- [Test-Time Prompt Tuning for Zero-Shot Generalization in Vision-Language Models](https://mlsanity.com/paper/2209.07511/) (2022)
- [RemoteCLIP: A Vision Language Foundation Model for Remote Sensing](https://mlsanity.com/paper/2306.11029/) (2023)
- [Tip-Adapter: Training-free Adaption of CLIP for Few-shot Classification](https://mlsanity.com/paper/2207.09519/) (2022)
- [Prompt-aligned Gradient for Prompt Tuning](https://mlsanity.com/paper/2205.14865/) (2022)
- [A Survey of Vision-Language Pre-Trained Models](https://mlsanity.com/paper/2202.10936/) (2022)
- [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)
- [CORA: Adapting CLIP for Open-Vocabulary Detection with Region Prompting and Anchor Pre-Matching](https://mlsanity.com/paper/2303.13076/) (2023)
- [Task Residual for Tuning Vision-Language Models](https://mlsanity.com/paper/2211.10277/) (2022)
- [C-TPT: Calibrated Test-Time Prompt Tuning for Vision-Language Models via Text Feature Dispersion](https://mlsanity.com/paper/2403.14119/) (2024)
- [CLIP goes 3D: Leveraging Prompt Tuning for Language Grounded 3D Recognition](https://mlsanity.com/paper/2303.11313/) (2023)
- [Dynamic Adapter Meets Prompt Tuning: Parameter-Efficient Transfer Learning for Point Cloud Analysis](https://mlsanity.com/paper/2403.01439/) (2024)
- [Dual Memory Networks: A Versatile Adaptation Approach for Vision-Language Models](https://mlsanity.com/paper/2403.17589/) (2024)
- [StitchFusion: Weaving Any Visual Modalities to Enhance Multimodal Semantic Segmentation](https://mlsanity.com/paper/2408.01343/) (2024)
- [Dual Prototype Evolving for Test-Time Generalization of Vision-Language Models](https://mlsanity.com/paper/2410.12790/) (2024)
- [Just Shift It: Test-Time Prototype Shifting for Zero-Shot Generalization with Vision-Language Models](https://mlsanity.com/paper/2403.12952/) (2024)

## Built on by (most cited first, 6 of 6)

- [CogVLA: Cognition-Aligned Vision-Language-Action Model via Instruction-Driven Routing & Sparsification](https://mlsanity.com/paper/2508.21046/) (2025)
- [Adapting Vision-Language Models Without Labels: A Comprehensive Survey](https://mlsanity.com/paper/2508.05547/) (2025)
- [Multi-Cache Enhanced Prototype Learning for Test-Time Generalization of Vision-Language Models](https://mlsanity.com/paper/2508.01225/) (2025)
- [Mitigating Long-Tail Bias in HOI Detection via Adaptive Diversity Cache](https://mlsanity.com/paper/2511.18811/) (2025)
- [Semantic-Aware Temporal Adaptation for UAV Anti-UAV Tracking](https://mlsanity.com/paper/2607.26511/) (2026)
- [On-Policy Distillation for Vision-Language Model Adaptation, an Effective Paradigm on Low-Quality Multimodal Data](https://mlsanity.com/paper/2609.10321/) (2026)

## Abstract (arXiv)

Test-time adaptation (TTA) is crucial in maintaining performance of Vision Language Models (VLMs) when facing distribution shifts, particularly when the source data or target labels are inaccessible. Existing TTA methods predominantly leverage the output probability distribution of CLIP for feature evaluation, resulting in biases under domain shifts, which cause misclassified features due to text priors or incorrect textual associations. To address these issues, we propose underlineBidirectional Prototype-Reward co-Evolution (BPRE), a novel VLMs framework with TTA that integrates feature quality assessment with prototype evolution via a synergistic feedback loop. First, the Multi-dimensional Quality-aware Reward Module (MQRM) is designed to evaluate feature quality and guide prototype refinement precisely. The continuous refinement of prototype quality via Prototype-Reward Interactive Evolution (PRIE) enhances the computation more robust. Through this bidirectional interaction, the precision of rewards and prototype evolution mutually reinforce each other, forming a self-evolving feedback cycle. Extensive experiments conducted on 15 diverse recognition datasets demonstrate that our model consistently achieves superior performance compared to other SOTA methods, and advances VLM generalization capabilities through emphasizing comprehensive feature evaluation.