diff --git a/units/en/unitbonus3/language-models.mdx b/units/en/unitbonus3/language-models.mdx index 9d873c3..db36cf7 100644 --- a/units/en/unitbonus3/language-models.mdx +++ b/units/en/unitbonus3/language-models.mdx @@ -20,9 +20,7 @@ There is therefore a potential synergy between LMs which can bring knowledge abo As a first attempt, the paper [“Grounding Large Language Models with Online Reinforcement Learning”](https://arxiv.org/abs/2302.02662v1) tackled the problem of **adapting or aligning a LM to a textual environment using PPO**. They showed that the knowledge encoded in the LM lead to a fast adaptation to the environment (opening avenue for sample efficiency RL agents) but also that such knowledge allowed the LM to better generalize to new tasks once aligned. - +