Investigating OCR-Sensitive Neurons to Improve Entity Recognition in Historical Documents

2024-09-25 · Emanuela Boros, Maud Ehrmann

This paper investigates the presence of OCR-sensitive neurons within the Transformer architecture and their influence on named entity recognition (NER) performance on historical documents. By analysing neuron activation patterns in response to clean and noisy text inputs, we identify and then neutralise OCR-sensitive neurons to improve model performance. Based on two open access large language models (Llama2 and Mistral), experiments demonstrate the existence of OCR-sensitive regions and show improvements in NER performance on historical newspapers and classical commentaries, highlighting the potential of targeted neuron modulation to improve models' performance on noisy text.

📄 PDF Abstract BibTeX arXiv:2409.16934

Code (1)

emanuelaboros/ocr-sensitive-neurons 공식 구현 pytorch

Tasks

named-entity-recognitionNamed Entity RecognitionNamed Entity Recognition (NER)NER Optical Character Recognition (OCR)

Methods 이 논문이 사용한 방법론

Attention 설명 없음

Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.

Position-Wise Feed-Forward Layer 설명 없음

Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…

BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…

Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…

Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…

Deep Neural Model Inspection and Comparison via Functional Neuron Pathways

2019-07-01 · ACL 2019 7 · James Fiacco, Samridhi Choudhary, Carolyn Rose

We introduce a general method for the interpretation and comparison of neural models. The method is used to factor a complex neural model into its functional components, which are comprised of sets of co-firing neurons t…

named-entity-recognition Named Entity Recognition Named Entity Recognition (NER)Natural Language Inference

Analysis of Named Entity Recognition and Linking for Tweets

2014-10-27 · Leon Derczynski, Diana Maynard, Giuseppe Rizzo, Marieke van Erp 외

Applying natural language processing for mining and intelligent information access to tweets (a form of microblog) is a challenging, emerging research area. Unlike carefully authored news text and other longer content, t…

Entity Disambiguation Language Identification named-entity-recognition Named Entity Recognition+2

Statistics of Visual Responses to Object Stimuli from Primate AIT Neurons to DNN Neurons

2016-12-12 · Qiulei Dong, Zhanyi Hu

Cadieu et al. (Cadieu,2014) reported that deep neural networks(DNNs) could rival the representation of primate inferotemporal cortex for object recognition. Lehky et al. (Lehky,2011) provided a statistical analysis on ne…

Object Object Recognition

Deep Convolutional Neural Networks in the Face of Caricature: Identity and Image Revealed

2018-12-28 · Matthew Q. Hill, Connor J. Parde, Carlos D. Castillo, Y. Ivette Colon 외

Real-world face recognition requires an ability to perceive the unique features of an individual face across multiple, variable images. The primate visual system solves the problem of image invariance using cascades of n…

Caricature Face Recognition Image-Variation Network Identification

Head Similarity: Modeling Structured Whole-Head Appearance Beyond Face Recognition

2026-05-08 · Yingfeng Wang, Yuxuan Xiao, Shengcai Liao arxiv

Many vision applications require identity consistency beyond strict biometric recognition, especially under non-frontal views or when facial cues are missing. However, conventional face recognition models enforce intra-i…

Face Recognition