In this note, we explore inference-time alignment through in-context
lea...
In-context learning (ICL) improves language models' performance on a var...
Language models (LMs) often struggle to pay enough attention to the inpu...
Large language models (LLMs) are increasingly adopted for a variety of t...
Large language models can perform new tasks in a zero-shot fashion, give...
Large pretrained language models have been performing increasingly well ...
Among the most critical limitations of deep learning NLP models are thei...
Modern toxic speech detectors are incompetent in recognizing disguised
o...
Modern deep learning models for NLP are notoriously opaque. This has
mot...
Understanding the dynamics of international politics is important yet
ch...
Contextualized word embeddings such as ELMo and BERT provide a foundatio...
Contextualized word embeddings such as ELMo and BERT provide a foundatio...
Wikipedia has a strong norm of writing in a 'neutral point of view' (NPO...