Latest Articles for Text Analysis

A new method identifies machine-generated texts by analyzing human writing styles.

Examining the impact of text reuse in historical and literary studies.

A method for handling ambiguities in proper names through regular polysemy.

A new approach enhances NLP models against adversarial attacks through targeted paraphrasing.

A comparison of ChatGPT and BERT in understanding word meaning shifts over time.

A new method for efficient data compression and retrieval.

New unbounded language model improves predictions using extensive data.

HQA-Attack creates high-quality adversarial examples in text while preserving meaning.

Examining methods to enhance ChatGPT's classification of implicit discourse relations.

This article examines how input length affects Large Language Models' reasoning skills.

A framework for updating ontologies with new ideas using language models.

New method improves sentence embeddings using multiple prompts for better accuracy.

This study presents a method to find meanings not listed in dictionaries.

A new method enhances resilience of models to adversarial examples through text prompt adjustment.

This article explores how paraphrasing improves language model performance in text classification.

A new method improves event argument extraction in complex documents.

A new method for assessing language processing tools shows promise for improvement.

A novel approach to enhance summarization skills in smaller models using larger models.

This study investigates the role of synthetic examples in authorship verification.

New method SSCAE improves adversarial example generation in natural language processing.

Two innovative algorithms effectively create subtle adversarial examples in language processing.

This study evaluates various language models for text classification using minimal training data.

Gecko offers an efficient model for processing and comparing texts effectively.

An automated method for aligning discourse relations from different labeling systems.

SnAG improves video grounding accuracy and efficiency for longer videos.

Innovative methods for training models using compressed text.

A new method enhances event linking in texts using AI-generated reasoning.

This article reviews methods for generating questions to improve event argument extraction.

LLM2Vec improves decoder-only models for better language processing.

Examining how similar subwords affect language model learning and performance.

An overview of tokenization's role in language processing.

A new approach using RGCNs improves sentiment classification from text data.

Introducing TrainlessGNN for faster node labeling in TAGs without traditional training methods.

This study analyzes language biases affecting gender and race in various texts.

A look at the progress in paraphrase generation methods using knowledge distillation.

New methods improve accuracy in linking objects and attributes in generated images.

A look at how to identify human and machine-written content.

A new method uses questions to identify event roles in documents.

New module QASE improves accuracy in machine reading comprehension tasks.

This study uses computational methods to analyze running titles in historical texts.