Latest Articles for Model Evaluation

A new benchmark aims to improve uncertainty assessment in language models.

A new method improves model reasoning through structured programming traces.

Examining how fine-tuning affects safety in language models across various tasks.

A fresh approach to evaluating ML models using Item Response Theory for better insights.

Strong baseline models enhance the evaluation of ML systems in healthcare.

A look at confidence intervals in few-shot learning and their impact on model evaluation.

Examining the understanding and output accuracy of language models.

Research highlights using influence functions to enhance PINN performance in physics problems.

A look into effective dimension and its impact on model training.

This paper evaluates how well language models explain scientific concepts.

This article examines GAMs as a solution for predictive performance and interpretability.

Examining how hard samples affect model performance and the reliability of test accuracy.

This article examines how different layers affect LLM performance.

Soft labels can improve machine learning model performance in uncertain data scenarios.

RepairBench sets benchmarks for comparing AI models in fixing software bugs.

This method enhances the reliability of language model confidence scores.

Learn how the applicability domain affects predictive model accuracy in various fields.

A method to estimate reliability of responses from large language models.

A new method for testing language models using randomized text.

A method to improve steering vector effectiveness in language models.

Explore the impact of shortcut learning on language models and their real-world applications.

This paper examines methods to compare generative models through embedding-based representations.

A framework to balance pseudo-label learning in machine learning.

New tool H-POPE improves accuracy of vision-language models.

A study on different models' abilities in In-Context Learning.

A new framework identifies when multimodal models use inappropriate training data.

This article discusses the need for transparency in language model benchmarks.

An overview of the strengths and flaws in today's Vision-Language Models.

A comprehensive study comparing methods for estimating confidence intervals in machine learning models.

A look at similarity networks to improve fairness in machine learning.

Learn strategies to improve model performance on imbalanced datasets.

A guide to understanding AI model performance using the FEET framework.

A framework for comparing forecasting models using principal components.

RLInspect helps analyze and improve reinforcement learning models effectively.

Examining how AI models handle text and images together.

Exploring how model size affects performance in OOD detection.

A new method enhances detection of unfamiliar data in deep learning models.

Are NLI tasks still relevant for testing large language models?

ICER framework tests safety measures in text-to-image models effectively.

A study reveals accuracy issues in AI-generated long texts.