Enregistré dans:
Détails bibliographiques
Auteurs principaux: Zeng, Fanwei, Miao, Changtao, Huang, Jing, Tan, Zhiya, Gong, Shutao, Yu, Xiaoming, Wang, Yang, Tan, Huazhe, Yao, Weibin, Li, Jianshu
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2512.21482
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866912789083193344
author Zeng, Fanwei
Miao, Changtao
Huang, Jing
Tan, Zhiya
Gong, Shutao
Yu, Xiaoming
Wang, Yang
Tan, Huazhe
Yao, Weibin
Li, Jianshu
author_facet Zeng, Fanwei
Miao, Changtao
Huang, Jing
Tan, Zhiya
Gong, Shutao
Yu, Xiaoming
Wang, Yang
Tan, Huazhe
Yao, Weibin
Li, Jianshu
contents Sophisticated text-centric forgeries, fueled by rapid AIGC advancements, pose a significant threat to societal security and information authenticity. Current methods for text-centric forgery analysis are often limited to coarse-grained visual analysis and lack the capacity for sophisticated reasoning. Moreover, they typically treat detection, grounding, and explanation as discrete sub-tasks, overlooking their intrinsic relationships for holistic performance enhancement. To address these challenges, we introduce LogicLens, a unified framework for Visual-Textual Co-reasoning that reformulates these objectives into a joint task. The deep reasoning of LogicLens is powered by our novel Cross-Cues-aware Chain of Thought (CCT) mechanism, which iteratively cross-validates visual cues against textual logic. To ensure robust alignment across all tasks, we further propose a weighted multi-task reward function for GRPO-based optimization. Complementing this framework, we first designed the PR$^2$ (Perceiver, Reasoner, Reviewer) pipeline, a hierarchical and iterative multi-agent system that generates high-quality, cognitively-aligned annotations. Then, we constructed RealText, a diverse dataset comprising 5,397 images with fine-grained annotations, including textual explanations, pixel-level segmentation, and authenticity labels for model training. Extensive experiments demonstrate the superiority of LogicLens across multiple benchmarks. In a zero-shot evaluation on T-IC13, it surpasses the specialized framework by 41.4% and GPT-4o by 23.4% in macro-average F1 score. Moreover, on the challenging dense-text T-SROIE dataset, it establishes a significant lead over other MLLM-based methods in mF1, CSS, and the macro-average F1. Our dataset, model, and code will be made publicly available.
format Preprint
id arxiv_https___arxiv_org_abs_2512_21482
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LogicLens: Visual-Logical Co-Reasoning for Text-Centric Forgery Analysis
Zeng, Fanwei
Miao, Changtao
Huang, Jing
Tan, Zhiya
Gong, Shutao
Yu, Xiaoming
Wang, Yang
Tan, Huazhe
Yao, Weibin
Li, Jianshu
Artificial Intelligence
Sophisticated text-centric forgeries, fueled by rapid AIGC advancements, pose a significant threat to societal security and information authenticity. Current methods for text-centric forgery analysis are often limited to coarse-grained visual analysis and lack the capacity for sophisticated reasoning. Moreover, they typically treat detection, grounding, and explanation as discrete sub-tasks, overlooking their intrinsic relationships for holistic performance enhancement. To address these challenges, we introduce LogicLens, a unified framework for Visual-Textual Co-reasoning that reformulates these objectives into a joint task. The deep reasoning of LogicLens is powered by our novel Cross-Cues-aware Chain of Thought (CCT) mechanism, which iteratively cross-validates visual cues against textual logic. To ensure robust alignment across all tasks, we further propose a weighted multi-task reward function for GRPO-based optimization. Complementing this framework, we first designed the PR$^2$ (Perceiver, Reasoner, Reviewer) pipeline, a hierarchical and iterative multi-agent system that generates high-quality, cognitively-aligned annotations. Then, we constructed RealText, a diverse dataset comprising 5,397 images with fine-grained annotations, including textual explanations, pixel-level segmentation, and authenticity labels for model training. Extensive experiments demonstrate the superiority of LogicLens across multiple benchmarks. In a zero-shot evaluation on T-IC13, it surpasses the specialized framework by 41.4% and GPT-4o by 23.4% in macro-average F1 score. Moreover, on the challenging dense-text T-SROIE dataset, it establishes a significant lead over other MLLM-based methods in mF1, CSS, and the macro-average F1. Our dataset, model, and code will be made publicly available.
title LogicLens: Visual-Logical Co-Reasoning for Text-Centric Forgery Analysis
topic Artificial Intelligence
url https://arxiv.org/abs/2512.21482