3 papers
cs.CL2023
ContextRef: Evaluating Referenceless Metrics For Image Description Generation
Elisa Kreiss, Eric Zelikman, Christopher Potts +1
Referenceless metrics (e.g., CLIPScore) use pretrained vision--language models to assess image descriptions directly without costly ground-truth reference texts. Such methods can f…
cs.CL2023
Context-VQA: Towards Context-Aware and Purposeful Visual Question Answering
Nandita Naik, Christopher Potts, Elisa Kreiss
Visual question answering (VQA) has the potential to make the Internet more accessible in an interactive way, allowing people who cannot see images to ask questions about them. How…
cs.HC2023
Characterizing Image Accessibility on Wikipedia across Languages
Elisa Kreiss, Krishna Srinivasan, Tiziano Piccardi +5
We make a first attempt to characterize image accessibility on Wikipedia across languages, present new experimental results that can inform efforts to assess description quality, a…