e-SNLI-VE
Emerging4papers using it
15HF downloads
1HF likes
2022first seen
The 'e-SNLI-VE' dataset is a benchmark that contains pairs of natural language explanations and visual inputs, used to evaluate the faithfulness of natural language explanations in vision-language tasks.
Papers using e-SNLI-VE (4)
- Probing Vision-Language Understanding through the Visual Entailment Task: promises and pitfallsNLKI: A Lightweight Natural Language Knowledge Integration Framework For Improving Small Vlms In Commonsense VQA TasksOn Advances in Text Generation from Images Beyond Captioning: A Case
Study in Self-RationalizationBenchmarking Faithfulness: Towards Accurate Natural Language
Explanations in Vision-Language Tasks