# Improving Applicability of Deep Learning based Token Classification models during Training

arXiv: https://arxiv.org/abs/2504.01028 · HTML: https://mlsanity.com/paper/2504.01028/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 35 references that are in the map.
- First posted on arXiv 28 Mar 2025.

## Builds on (most cited first, 22 of 22)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [HuggingFace's Transformers: State-of-the-art Natural Language Processing](https://mlsanity.com/paper/1910.03771/) (2019)
- [Retrieval-Augmented Generation for Large Language Models: A Survey](https://mlsanity.com/paper/2312.10997/) (2023)
- [LayoutLM: Pre-training of Text and Layout for Document Image Understanding](https://mlsanity.com/paper/1912.13318/) (2019)
- [An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition](https://mlsanity.com/paper/1507.05717/) (2015)
- [LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking](https://mlsanity.com/paper/2204.08387/) (2022)
- [LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding](https://mlsanity.com/paper/2012.14740/) (2020)
- [OCR-free Document Understanding Transformer](https://mlsanity.com/paper/2111.15664/) (2021)
- [End-to-end Sequence Labeling via Bi-directional LSTM-CNNs-CRF](https://mlsanity.com/paper/1603.01354/) (2016)
- [Large Language Models for Information Retrieval: A Survey](https://mlsanity.com/paper/2308.07107/) (2023)
- [Character Region Awareness for Text Detection](https://mlsanity.com/paper/1904.01941/) (2019)
- [ICDAR2019 Competition on Scanned Receipt OCR and Information Extraction](https://mlsanity.com/paper/2103.10213/) (2021)
- [Unified Structure Generation for Universal Information Extraction](https://mlsanity.com/paper/2203.12277/) (2022)
- [Large Language Models for Generative Information Extraction: A Survey](https://mlsanity.com/paper/2312.17617/) (2023)
- [Unifying Multimodal Retrieval via Document Screenshot Embedding](https://mlsanity.com/paper/2406.11251/) (2024)
- [BROS: A Pre-trained Language Model Focusing on Text and Layout for Better Key Information Extraction from Documents](https://mlsanity.com/paper/2108.04539/) (2021)
- [Graph Convolution for Multimodal Information Extraction from Visually Rich Documents](https://mlsanity.com/paper/1903.11279/) (2019)
- [TRIE: End-to-End Text Reading and Information Extraction for Document Understanding](https://mlsanity.com/paper/2005.13118/) (2020)
- [Cross-Modal Retrieval: A Systematic Review of Methods and Future Directions](https://mlsanity.com/paper/2308.14263/) (2023)
- [Generative Cross-Modal Retrieval: Memorizing Images in Multimodal Language Models for Retrieval and Beyond](https://mlsanity.com/paper/2402.10805/) (2024)
- [Generative Multi-Modal Knowledge Retrieval with Large Language Models](https://mlsanity.com/paper/2401.08206/) (2024)
- [Enhancement of the e-Invoicing Systems by Increasing the Efficiency of Workflows via Disruptive Technologies](https://mlsanity.com/paper/2010.07636/) (2020)

## Built on by (most cited first, 1 of 1)

- [Beyond RAG for Cyber Threat Intelligence: A Systematic Evaluation of Graph-Based and Agentic Retrieval](https://mlsanity.com/paper/2604.11419/) (2026)

## Abstract (arXiv)

This paper shows that further evaluation metrics during model training are needed to decide about its applicability in inference. As an example, a LayoutLM-based model is trained for token classification in documents. The documents are German receipts. We show that conventional classification metrics, represented by the F1-Score in our experiments, are insufficient for evaluating the applicability of machine learning models in practice. To address this problem, we introduce a novel metric, Document Integrity Precision (DIP), as a solution for visual document understanding and the token classification task. To the best of our knowledge, nothing comparable has been introduced in this context. DIP is a rigorous metric, describing how many documents of the test dataset require manual interventions. It enables AI researchers and software developers to conduct an in-depth investigation of the level of process automation in business software. In order to validate DIP, we conduct experiments with our created models to highlight and analyze the impact and relevance of DIP to evaluate if the model should be deployed or not in different training settings. Our results demonstrate that existing metrics barely change for isolated model impairments, whereas DIP indicates that the model requires substantial human interventions in deployment. The larger the set of entities being predicted, the less sensitive conventional metrics are, entailing poor automation quality. DIP, in contrast, remains a single value to be interpreted for entire entity sets. This highlights the importance of having metrics that focus on the business task for model training in production. Since DIP is created for the token classification task, more research is needed to find suitable metrics for other training tasks.