In LaTeX text recognition using Transformer-based architectures, this paper
identifies certain "bias" issues. For instance, $e-t$ is frequently
misrecognized as $e^{-t}$. This bias stems from the inherent characteristics of
the dataset. To mitigate this bias, we propose a LaTeX printed text recognition
model trained on a mixed dataset of pseudo-formulas and pseudo-text. The model
employs a Swin Transformer as the encoder and a RoBERTa model as the decoder.
Experimental results demonstrate that this approach reduces "bias", enhancing
the accuracy and robustness of text recognition. For clear images, the model
strictly adheres to the image content; for blurred images, it integrates both
image and contextual information to produce reasonable recognition results.

使用基于 Transformer 的架构进行 LaTeX 文本识别，识别到存在的 “偏差” 问题并提出使用混合数据集训练的 LaTeX 打印文本识别模型，该模型在编码器中采用 Swin Transformer，解码器中采用 RoBERTa 模型。实验结果表明，该方法减小了 “偏差”，提高了文本识别的准确性和鲁棒性。对于清晰图像，模型严格遵循图像内容；对于模糊图像，它整合图像和上下文信息以产生合理的识别结果。