Papers by Wei Bin Au Yeong

1 papers
ROME: Evaluating Pre-trained Vision-Language Models on Reasoning beyond Visual Common Sense (2023.findings-emnlp)

Copied to clipboard

Challenge: a vision-language model with commonsense knowledge can reason beyond common sense . however, pre-trained vision-linguistic models are incapable of interpreting counter-intuitive content .
Approach: They introduce a probing dataset to evaluate vision-language models' reasoning abilities . they use images that defy commonsense knowledge to test their reasoning abilities.
Outcome: The proposed dataset evaluates whether pre-trained vision-language models can reason beyond common sense . it contains images that defy commonsense knowledge with regards to color, shape, material, size and position .

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations