NLP Pipeline

Natural Language Processing (NLP) pipelines are sequences of computational steps designed to extract meaning and insights from textual data. Current research focuses on improving pipeline components like tokenization (adapting to linguistic nuances and high internal complexity), named entity recognition (especially in specialized domains like legal texts and social media), and embedding generation (enhancing numerical and contextual understanding). These advancements are crucial for various applications, including legal support, educational technology, and autonomous vehicle safety, by enabling more accurate and efficient analysis of large text corpora.

Papers

April 2, 2024

Laying Anchors: Semantically Priming Numerals in Language Modeling
Mandar Sharma, Rutuja Murlidhar Taware, Pravesh Koirala, Nikhil Muralidhar, Naren Ramakrishnan
Pre Trained Language Model Semantics Surfaced Learned Embeddings Thinking Token NLP Pipeline Numeral System Internal Numeracy

October 23, 2023

Statistical Depth for Ranking and Characterizing Transformer-Based Text Embeddings
Parker Seegmiller, Sarah Masud Preum
Text Embeddings Partial Ranking Text Transformation NLP Pipeline Depth Function Statistical Depth

August 24, 2023

Advancing Hungarian Text Processing with HuSpaCy: Efficient and Accurate NLP Pipelines
György Orosz, Gergő Szabó, Péter Berkecz, Zsolt Szántó, Richárd Farkas
Entity Recognition Natural Language Processing Tool Morphological Annotation NLP Pipeline Hungarian Corpus Text Mining Model

July 14, 2023

MorphPiece : A Linguistic Tokenizer for Large Language Models
Haris Jabbar
Online Tokenizer Morphological Tokenization NLP Pipeline

May 24, 2023

Automated Refugee Case Analysis: An NLP Pipeline for Supporting Legal Practitioners
Claire Barale, Michael Rovatsos, Nehal Bhuta
Named Entity Recognition Contextual Embeddings Legal Education NLP Pipeline Legal Analysis

May 9, 2023

Estimating related words computationally using language model from the Mahabharata -- an Indian epic
Vrunda Gadesha, Keyur D Joshi, Shefali Naik
Language Model Natural Language Processing Text Analysis Social Medium Sentiment NLP Pipeline Converting Epic

January 11, 2023

The Role of Interactive Visualization in Explaining (Large) NLP Models: from Data to Inference
Richard Brath, Daniel Keim, Johannes Knittel, Shimei Pan, Pia Sommerauer, Hendrik Strobelt
Raw Data Line by Line Explanation Scientific Inference NLP Model Interactive Visualization Modern Language Model NLP Pipeline

October 18, 2022

Graph-Based Multilingual Label Propagation for Low-Resource Part-of-Speech Tagging
Ayyoob Imani, Silvia Severini, Masoud Jalili Sabet, François Yvon, Hinrich Schütze
Label Propagation Part of Speech NLP Pipeline

May 24, 2022

K-12BERT: BERT for K-12 education
Vasu Goel, Dhruv Sahnan, Venktesh V, Gaurav Sharma, Deep Dwivedi, Mukesh Mohania
Language Model Pre Trained Language Model Ticket BERT NLP Pipeline Hierarchical Taxonomy

March 21, 2022

Neural Token Segmentation for High Token-Internal Complexity
Idan Brusilovsky, Reut Tsarfaty
Complexity Level Token Representation NLP Pipeline Class Token

January 18, 2022

Annotating the Tweebank Corpus on Named Entity Recognition and Building NLP Models for Social Media Analysis
Hang Jiang, Yining Hua, Doug Beeferman, Deb Roy
Entity Recognition Social Medium Named Entity Recognition NLP Model Annotation Rather NLP Pipeline Twitter Corpus

November 5, 2021

Disengagement Cause-and-Effect Relationships Extraction Using an NLP Pipeline
Yangtao Zhang, X. Jessie Yang, Feng Zhou
Autonomous Driving Autonomous Vehicle NLP Pipeline Causal Relation Extraction Disengagement Report

NLP Pipeline

Papers

Laying Anchors: Semantically Priming Numerals in Language Modeling

Statistical Depth for Ranking and Characterizing Transformer-Based Text Embeddings

Advancing Hungarian Text Processing with HuSpaCy: Efficient and Accurate NLP Pipelines

MorphPiece : A Linguistic Tokenizer for Large Language Models

Automated Refugee Case Analysis: An NLP Pipeline for Supporting Legal Practitioners

Estimating related words computationally using language model from the Mahabharata -- an Indian epic

The Role of Interactive Visualization in Explaining (Large) NLP Models: from Data to Inference

Graph-Based Multilingual Label Propagation for Low-Resource Part-of-Speech Tagging

K-12BERT: BERT for K-12 education

Neural Token Segmentation for High Token-Internal Complexity

Annotating the Tweebank Corpus on Named Entity Recognition and Building NLP Models for Social Media Analysis

Disengagement Cause-and-Effect Relationships Extraction Using an NLP Pipeline