EPFL Data Science & AI Lab (dlab)

AI safety · alignment · language models

The Data Science & AI Lab at EPFL, led by Robert West, works on AI safety and alignment: how language models acquire values, how those values can be installed early enough to hold, and how to tell whether they actually did. Alongside that we study what models represent internally and how to make them more efficient. This Hub hosts the models, datasets, and benchmarks from our public releases.

Research projects

Synthetic Persona Pretraining has its own organization — @dlab-spp — with all models, the pretraining and post-training data, and the evaluation benchmarks.

Browse all GitHub repositories or visit the lab website.