Technology ❯ Data Science

Benchmarking

Performance Evaluation Model Evaluation Evaluation Metrics Performance Metrics Evaluation Methods Dataset Creation Standardized Datasets Factuality Benchmarks Dataset Curation

ArXiv Preprint Says Prompts Drive Many LVLM Hallucinations

The authors present tools to test and reduce instruction-induced mistakes in image-text models.