# Implicit Language Model in LSTM for OCR

arXiv: https://arxiv.org/abs/1805.09441 · HTML: https://mlsanity.com/paper/1805.09441/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 23 May 2018.

## Built on by (most cited first, 6 of 6)

- [Evaluation of HTR models without Ground Truth Material](https://mlsanity.com/paper/2201.06170/) (2022)
- [Self-Training of Handwritten Word Recognition for Synthetic-to-Real Adaptation](https://mlsanity.com/paper/2206.03149/) (2022)
- [Handwritten Word Recognition using Deep Learning Approach: A Novel Way of Generating Handwritten Words](https://mlsanity.com/paper/2303.07514/) (2023)
- [ShadowGenes: Leveraging Recurring Patterns within Computational Graphs for Model Genealogy](https://mlsanity.com/paper/2501.11830/) (2025)
- [OCR Language Models with Custom Vocabularies](https://mlsanity.com/paper/2308.09671/) (2023)
- [A Lightweight Multi-Module Fusion Approach for Korean Character Recognition](https://mlsanity.com/paper/2504.05770/) (2025)

## Abstract (arXiv)

Neural networks have become the technique of choice for OCR, but many aspects of how and why they deliver superior performance are still unknown. One key difference between current neural network techniques using LSTMs and the previous state-of-the-art HMM systems is that HMM systems have a strong independence assumption. In comparison LSTMs have no explicit constraints on the amount of context that can be considered during decoding. In this paper we show that they learn an implicit LM and attempt to characterize the strength of the LM in terms of equivalent n-gram context. We show that this implicitly learned language model provides a 2.4% CER improvement on our synthetic test set when compared against a test set of random characters (i.e. not naturally occurring sequences), and that the LSTM learns to use up to 5 characters of context (which is roughly 88 frames in our configuration). We believe that this is the first ever attempt at characterizing the strength of the implicit LM in LSTM based OCR systems.