Selective ‘selective prediction’: reducing unnecessary abstention in vision-language reasoning
Read: ACL Findings
Research
Insights from the cutting-edge
Many of our team have deep roots in AI research. We’re excited to share our latest findings with the research community.
Papers
When does monolingual data help multilingual translation: the role of domain and model scale
Read: NAACL
Lost in the middle: how language models use long contexts
Read: TACL
Deal, or no deal (or who knows)? Forecasting uncertainty in conversations using large language models
Read: ACL Findings
Improving Wikipedia verifiability with AI
Read: Nature Machine Intelligence
Cross-domain image captioning with discriminative finetuning
Read: CVPR
Can discrete information extraction prompts generalize across language models?
Read: ICLR