Latest Articles for Data Evaluation

Combining human reviewers with LLMs improves biomedical research evaluations.

A challenge focusing on deep generative models for realistic medical image generation.

A new system for assessing language models using real-world data streams.

A new method to assess commonsense reasoning in AI models through open-ended tasks.

New GAIA dataset sheds light on action quality in AI-generated content.

A new method to assess generative models with minimal data generation.

A new benchmark tests compositional reasoning in advanced models.

New dataset helps assess AI text accuracy and reliability.

A new benchmark evaluates how language models handle text changes.

A toolkit for assessing performance of retrieval-augmented models in specific domains.

VideoVista offers a comprehensive evaluation for video question-answering models.

Methods for measuring treatment effects across diverse groups and time frames.

This article presents a new method for assessing text-to-image models effectively.

Dysca introduces a new way to assess LVLM performance using synthetic data.

A new method measures how language models adapt their beliefs with new evidence.

A new benchmark to assess AI agents' performance in biomedical literature and knowledge graphs.

Introducing FairMedFM to evaluate the fairness of foundation models in healthcare.

This study assesses how medical LVLMs perform amidst hallucinations using a new dataset.

Exploring machine learning models and new datasets for improved security.

FKEA offers a fresh way to assess generative models without needing reference datasets.

A look at the benefits of segment-level evaluation methods for translation quality.

New metrics and EdgeHead module enhance 3D detection for autonomous vehicles.

A new approach enhances the accuracy of language model evaluations.

Improving how models handle evidence in long documents builds user trust.

BiasAlert enhances bias detection in language models for fairer AI outputs.

A new method to assess accuracy in language model outputs.

A new benchmark sheds light on hallucination in vision language models.

This study highlights the importance of dataset granularity in improving image-text retrieval systems.

Introducing an efficient way to evaluate the quality of generated samples using latent density scores.

A new benchmark improves models' understanding of long videos and language.

HaloQuest addresses hallucination issues in vision-language models with a new dataset.

A new benchmark seeks to enhance evaluations of OIE systems for better performance insights.

A new benchmark to test visual-language models on minimal changes in images and captions.

This study highlights the need for LLMs to know when to abstain.

Proper scoring rules enhance the evaluation of probabilistic forecasts across various fields.

A framework for better estimating treatment effects in paired cluster-randomized experiments.

Using AI-generated relevance marks for efficient evaluation of information retrieval systems.

A new method improves evaluation accuracy in authorship verification by reducing topic leakage.

A new framework enhances evaluation of RAG systems in specialized domains.

New methods offer better evaluation of language understanding in models.