T2T-VICL: Unlocking the Boundaries of Cross-Task Visual In-Context Learning via Implicit Text-Driven VLMs

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Xia, Shao-Jun, Zhang, Huixin, Tu, Zhengzhong
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915741972824064
author Xia, Shao-Jun
Zhang, Huixin
Tu, Zhengzhong
author_facet Xia, Shao-Jun
Zhang, Huixin
Tu, Zhengzhong
contents In large language models (LLM), in-context learning (ICL) refers to performing new tasks by conditioning on small demonstrations provided in the input context. Recent advances in visual in-context learning (VICL) demonstrate promising capabilities for solving downstream tasks by unified vision-language models (VLMs). When the visual prompt and the target images originate from different visual tasks, can VLMs still enable VICL? In the paper, we propose a fully collaborative pipeline, i.e. T2T-VICL, for VLMs to investigate the potential of cross-task VICL. Fundamentally, we design a mechanism to generate and select text prompts that best implicitly describe the differences between two distinct low-level vision tasks, and construct the first cross-task VICL dataset. Building upon this, we propose a novel inference framework that combines perceptual score-based reasoning with traditional evaluation metrics to perform cross-task VICL. Our approach achieves top-tier results across twelve cross-task scenarios and second-tier performance in nine additional scenarios, unlocking the boundaries of cross-task VICL within VLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2511_16107
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle T2T-VICL: Unlocking the Boundaries of Cross-Task Visual In-Context Learning via Implicit Text-Driven VLMs
Xia, Shao-Jun
Zhang, Huixin
Tu, Zhengzhong
Computer Vision and Pattern Recognition
Artificial Intelligence
In large language models (LLM), in-context learning (ICL) refers to performing new tasks by conditioning on small demonstrations provided in the input context. Recent advances in visual in-context learning (VICL) demonstrate promising capabilities for solving downstream tasks by unified vision-language models (VLMs). When the visual prompt and the target images originate from different visual tasks, can VLMs still enable VICL? In the paper, we propose a fully collaborative pipeline, i.e. T2T-VICL, for VLMs to investigate the potential of cross-task VICL. Fundamentally, we design a mechanism to generate and select text prompts that best implicitly describe the differences between two distinct low-level vision tasks, and construct the first cross-task VICL dataset. Building upon this, we propose a novel inference framework that combines perceptual score-based reasoning with traditional evaluation metrics to perform cross-task VICL. Our approach achieves top-tier results across twelve cross-task scenarios and second-tier performance in nine additional scenarios, unlocking the boundaries of cross-task VICL within VLMs.
title T2T-VICL: Unlocking the Boundaries of Cross-Task Visual In-Context Learning via Implicit Text-Driven VLMs
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2511.16107