Latest Articles for Evaluation

DEnsity offers a fresh approach to evaluate dialogue systems based on human conversation patterns.

This article discusses the benefits of using diverse user feedback for better recommendations.

ArgU creates structured arguments based on factual information for effective discussions.

This study assesses GPT-3's ability to summarize medical literature effectively.

A mathematical method to evaluate the beauty of music performances.

This study evaluates periodontal care in Brazilian Dental Specialty Centers.

This article examines the effectiveness of AI-generated explanations for users.

A competition to improve automated Foley sound creation for multimedia.

C-Eval assesses reasoning and knowledge skills of LLMs in the Chinese language.

A new dataset improves how machines read and respond to documents.

An analysis of the RACE dataset's strengths and weaknesses for reading comprehension.

A critical look at language model benchmarks and their implications for human performance.

This article presents a new method for handling missing scores in NLP system evaluations.

Learn how chatbots are being trained to respond with empathy.

mLongT5 efficiently manages longer texts across multiple languages.

A new method enhances how we evaluate AI-generated images from text descriptions.

A study on creating structured instructions through hierarchical task decomposition.

IKDSumm effectively summarizes tweets during disasters using disaster-specific knowledge.

A new taxonomy to improve LLM performance on complex tasks.

A new method to assess argument quality by considering context.

Study assesses methods for evaluating language models in understanding language.

Seahorse provides a large collection of multilingual summaries with human ratings.

Research advancements in translating cultural references using machine translation systems.

A new method to integrate various medical data types for better analysis.

Assessing language models' performance across various human demographics is crucial for effective usage.

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

Introducing a structured framework for effective reasoning over long texts.

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

This study compares social norms between Chinese and American cultures through data analysis.

A new approach to summarizing tables based on user questions for better insights.

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

A new dataset improves language models' ability to understand instructions across various languages.

A new method addresses the challenges faced by language models in providing accurate answers.

A method to assess abstaining classifiers by estimating their missing predictions.

Clarification questions are essential for effective communication in conversational systems.

A new method improves video summarization for sign language content.

Enhancing model capabilities for linking various data types effectively.

A tool to assess large language models' multi-step reasoning capabilities.

Combining reference-based and reference-free methods for better summarization assessment.

Study shows LLMs provide more natural translations, especially for idiomatic phrases.