Parents:
Trust in NLP
Related Tags:
ExpandDescendants
2 Documents (Short List) 
Anthropic sur Twitter : "We examine which safety techniques for LMs are more robust to human-written, adversarial inputs ..."
Tags:
2022-08-25 About
[2006.09462] Selective Question Answering under Domain Shift
Tags:
2020-06-30 About
Properties