A Multimodal Framework for Financial Fake News Detection for Brazilian Portuguese

2026-04-12 · Source: Paper Index on ACL Anthology · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics · Depth: Expert, quick

Summary

A new multimodal framework has been developed to detect financial fake news in Brazilian Portuguese, addressing the common oversight of image-based content in existing detection methods. The system combines Natural Language Processing (NLP) with an image-to-text classification strategy. It utilizes Tesseract-based Optical Character Recognition (OCR) to extract text from images, which is then processed through the same unified pipeline used for textual content classification. Experiments conducted on the Fake.BR, FakeRecogna corpus, and BBC News Brasil datasets demonstrated that this approach achieved 98% accuracy when using BERTimbau fine-tuned on financial news. These results highlight the critical importance of analyzing visual text and confirm the effectiveness of a multimodal strategy for disinformation detection.

Key takeaway

For research scientists developing disinformation detection systems, you should consider incorporating multimodal analysis, specifically extracting and processing text from images. This approach, demonstrated to achieve 98% accuracy with BERTimbau fine-tuned on financial news, significantly enhances detection capabilities by addressing the often-neglected visual dimension of fake news, particularly in languages like Brazilian Portuguese.

Key insights

Multimodal analysis, integrating text and image-extracted text, significantly improves financial fake news detection.

Principles

• Visual text is critical for disinformation detection.
• Multimodal strategies enhance accuracy.

Method

The system extracts text from images using Tesseract-based OCR, then processes it with NLP alongside original text using a unified classification pipeline.

In practice

• Integrate OCR for image-based text analysis.
• Fine-tune BERTimbau for financial news.

Topics

• Financial Fake News Detection
• Multimodal Framework
• Brazilian Portuguese
• Natural Language Processing
• Optical Character Recognition

Best for: Research Scientist, AI Scientist, Machine Learning Engineer, NLP Engineer

Related on AIssential

• Exploring Knowledge Graphs for Automatic Fake News Detection in Portuguese — Knowledge Graphs offer an explainable method for automated fake news detection by modeling news as factual events.
• AI system spots fake reviews by combining text, images and user behavior — Multimodal AI combining text, images, and behavior significantly improves fake e-commerce review detection.
• Lexical and Syntactic Characterization of Fake News in Portuguese Produced by Humans and by Machines — Machine-generated Portuguese fake news shows distinct lexical and syntactic patterns compared to human-produced content.
• Misinformation Span Detection in Videos via Audio Transcripts — Pinpointing misinformation within videos via audio transcript spans offers more granular, interpretable detection than video-level classification.
• KITE: A Tri-Modal Transformer Integrating Text, Images, and Knowledge Graphs for Fake News Detection — KITE integrates text, images, and knowledge graphs via a tri-modal transformer for enhanced fake news detection and interpretability.
• On GATE, Text and Social Media Analysis, and Detecting Misinformation Online — A multi-layer analytical pipeline effectively extracts and categorizes digital identity aspects of agrifood products from online discourse.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Paper Index on ACL Anthology.