Test-Time Visual In-Context Tuning

2025-03-27 · CVPR 2025 1 · Jiahao Xie, Alessio Tonioni, Nathalie Rauschmayr, Federico Tombari, Bernt Schiele

Visual in-context learning (VICL), as a new paradigm in computer vision, allows the model to rapidly adapt to various tasks with only a handful of prompts and examples. While effective, the existing VICL paradigm exhibits poor generalizability under distribution shifts. In this work, we propose test-time Visual In-Context Tuning (VICT), a method that can adapt VICL models on the fly with a single test sample. Specifically, we flip the role between the task prompts and the test sample and use a cycle consistency loss to reconstruct the original task prompt output. Our key insight is that a model should be aware of a new test distribution if it can successfully recover the original task prompts. Extensive experiments on six representative vision tasks ranging from high-level visual understanding to low-level image processing, with 15 common corruptions, demonstrate that our VICT can improve the generalizability of VICL to unseen new domains. In addition, we show the potential of applying VICT for unseen tasks at test time. Code: https://github.com/Jiahao000/VICT.

📄 PDF Abstract BibTeX arXiv:2503.21777

Code (1)

jiahao000/vict 공식 구현

Tasks

In-Context Learning

Methods 이 논문이 사용한 방법론

AWARE We propose to theoretically and empirically examine the effect of incorporating weighting schemes into walk-aggregating GNNs. To this end, we propose a simple, interpretable, and…

FLIP https://developer.nvidia.com/blog/flip-a-difference-evaluator-for-alternating-images/

Cycle Consistency Loss Cycle Consistency Loss is a type of loss used for generative adversarial networks that performs unpaired image-to-image translation. It was introduced with the…

VisRef: Visual Refocusing while Thinking Improves Test-Time Scaling in Multi-Modal Large Reasoning Models

2026-02-27 · Soumya Suvra Ghosal, Youngeun Kim, Zhuowei Li, Ritwick Chaudhry 외 arxiv

Advances in large reasoning models have shown strong performance on complex reasoning tasks by scaling test-time compute through extended reasoning. However, recent studies observe that in vision-dependent tasks, extende…

Reinforcement Learning Visual Reasoning

In-context Prompt Learning for Test-time Vision Recognition with Frozen Vision-language Model

2024-03-10 · Junhui Yin, Xinyu Zhang, Lin Wu, Xiaojie Wang

Current pre-trained vision-language models, such as CLIP, have demonstrated remarkable zero-shot generalization capabilities across various downstream tasks. However, their performance significantly degrades when test in…

In-Context Learning Language Modeling Language Modelling Prompt Learning+1

Thinking in Frames: How Visual Context and Test-Time Scaling Empower Video Reasoning

2026-01-28 · Chengzu Li, Zanyi Wang, Jiaang Li, Yi Xu 외 arxiv

Vision-Language Models have excelled at textual reasoning, but they often struggle with fine-grained spatial understanding and continuous action planning, failing to simulate the dynamics required for complex visual reas…

Zero-shot Generalization Visual Reasoning Video Generation

Personalized Vision via Visual In-Context Learning

2025-09-29 · Yuxin Jiang, Yuchao Gu, Yiren Song, Ivor Tsang 외 arxiv

Modern vision models, trained on large-scale annotated datasets, excel at predefined tasks but struggle with personalized vision -- tasks defined at test time by users with customized objects or novel objectives. Existin…

Multi-Perspective Transformers in ARC-AGI-2 Challenge

2026-05-01 · Caleb Talley, Vedant Tibrewal, Seun Adekunle, Weiwen Dong 외 arxiv

ARC-AGI-2 is a benchmark of human-intuitive visual puzzles that measures a machine's ability to generalize from limited examples, interpret symbolic meaning, and flexibly apply rules in varying contexts. In this paper, w…