Multimodal large language models (MLLMs) are increasingly used for OCR in document and knowledge-processing pipelines, yet their ability to faithfully read real handwriting remains underexplored. E...
<p><strong>Tác giả:</strong> Zinuo Guo, Min Zhang, Bo Jiang</p>
<p>Multimodal large language models (MLLMs) are increasingly used for OCR in document and knowledge-processing pipelines, yet their ability to faithfully read real handwriting remains underexplored. Existing benchmarks emphasize printed text or clean single-line inputs and provide limited coverage of multilingual handwriting, writer errors, and structurally complex mathematical expressions. We introduce OmniHandwritingOCR, a diagnostic benchmark for evaluating MLLM/OCR systems on handwritten OCR...</p>
<p>📄 DOI: <a href="https://doi.org/https://doi.org/10.1145/3799682.3840823" target="_blank" rel="noopener">https://doi.org/10.1145/3799682.3840823</a></p>
<p>📖 <a href="https://doi.org/https://doi.org/10.1145/3799682.3840823" target="_blank" rel="noopener">Xem bài báo đầy đủ</a></p>
Bài viết được biên soạn lại bởi LIFVietnam. Nguồn tham khảo: OpenAlex
