Latest Articles for Natural Language Processing

Introducing a dataset for better tracking of entity states in procedural texts.

A new approach for better example selection in language models.

A new method streamlines data extraction from complex tables using schema-driven techniques.

This study assesses ChatGPT and OpenAssistant's effectiveness in classifying social media data.

Examines issues of consistency in large language models and their implications.

Learn how OIE enhances event detection from rich to low-resource areas.

A study highlighting the benefits of grammar constraints in language model outputs.

X-LLM integrates images, videos, and speech into language processing.

This study examines how prompt placement affects language model performance.

A new self-supervised method enhances logical reasoning without extensive labeled data.

Examining how LLMs handle questions without clear answers.

A new framework enhances question-knowledge alignment for accurate language model responses.

This article explores how training data affects image generation by text-to-image models.

A method to enhance black-box language models using smaller, trainable models.

TACR model improves how systems answer questions using tables and text.

Utilizing annotator differences can enhance natural language processing accuracy.

A new framework improves analysis of biomedical literature by linking entities and extracting events.

A new system enhances accuracy in answering various questions through specialized models.

A study reveals limitations in retrieval-augmented language models for text generation.

EXnet offers innovative solutions for text classification tasks with unlimited examples.

LiPoR offers a new way to handle plausible explanations without clear labels.

Combining active learning and parameter-efficient fine-tuning improves model performance with limited data.

Introducing a structured framework for effective reasoning over long texts.

Research highlights the importance of representation consistency in NLP models.

Examining the issues of inaccurate outputs in large language models.

MMSMR dataset aims to improve chatbot conversation evaluation with diverse human responses.

This study evaluates language models for coreference resolution without prior examples.

LMSI allows language models to enhance performance without extensive human input.

A new approach enhances models' ability to compare different entities using diverse data.

This study examines how language models respond to errors in text.

Research improves OpenQA models' ability to handle slight question changes.

Dynosaur automates instruction-tuning data creation, offering quality and cost benefits.

New methods improve finetuning of large models using less memory.

WebIE tackles the challenges of extracting structured data from web content.

This study evaluates how LLMs enhance training data for commonsense reasoning across languages.

Introducing a system that explains the evaluation of machine-generated text clearly.

This article discusses enhancing few-shot learning through effective example selection.

A new method enhances decision-making for large language models in web navigation.

New methods improve machine reasoning and explanation without labeled data.

Combining language models and Wikidata improves accurate question answering.