TexTAR : Textual Attribute Recognition in Multi-domain and Multi-lingual Document Images

2025-09-16

AI Summary

Addressing the dual challenges of low computational efficiency and poor noise robustness in text attribute recognition (e.g., bold, italic, underline, strikethrough) for multi-domain, multilingual document images, this paper proposes a context-aware multi-task Transformer architecture. Methodologically, it introduces (1) 2D Rotary Position Embedding (2D RoPE) to explicitly model the two-dimensional spatial layout of text; (2) a context-aware data selection pipeline to enhance training sample relevance; and (3) MMTAD—the first large-scale, multilingual, multi-domain dataset with fine-grained text attribute annotations. Evaluated on cross-lingual and cross-domain benchmarks, our approach significantly outperforms existing state-of-the-art methods. Results demonstrate that explicit 2D positional modeling and context-aware multi-task learning are critical for improving both accuracy and generalization in text attribute recognition.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Recognizing textual attributes such as bold, italic, underline and strikeout is essential for understanding text semantics, structure, and visual presentation. These attributes highlight key information, making them crucial for document analysis. Existing methods struggle with computational efficiency or adaptability in noisy, multilingual settings. To address this, we introduce TexTAR, a multi-task, context-aware Transformer for Textual Attribute Recognition (TAR). Our novel data selection pipeline enhances context awareness, and our architecture employs a 2D RoPE (Rotary Positional Embedding)-style mechanism to incorporate input context for more accurate attribute predictions. We also introduce MMTAD, a diverse, multilingual, multi-domain dataset annotated with text attributes across real-world documents such as legal records, notices, and textbooks. Extensive evaluations show TexTAR outperforms existing methods, demonstrating that contextual awareness contributes to state-of-the-art TAR performance.

Problem

• Recognizing textual attributes in multilingual document images
• Addressing computational inefficiency in noisy multilingual settings
• Improving adaptability for multi-domain document analysis

Innovation

• Transformer-based multi-task attribute recognition
• 2D RoPE mechanism for context integration
• Multilingual multi-domain dataset MMTAD