Kyle Lo

research

I was formerly a research scientist at the Allen Institute for AI, where I co-led the OLMo project on open language modeling. I’m now working on something new.

My current research focuses on large-scale pretraining of language models, with an emphasis on data curation, data mixing, benchmark validity, efficient experimentation, and generalization across model scales. I also like building useful data and software artifacts that support research, like large-scale corpora and OCR tools. Previously, I’ve worked on specializing language models to domains, human-AI interaction for sensemaking over large collections, and augmented reading interfaces.