IRL with noisy observations?

I have a question about inverse reinforcement learning. As far as my research went, I found papers that deal with noisy observations of the state, but that’s what the agent sees and the data scientist observes as well (that is, the agent measures the current state, i.e. S, with error and the data scientist has access to that observation as well).

But what happens if the information sets are different? What if the agent observes a noisy measure of S, but the data scientist knows S for sure (as if it were ex-post) and does not have access to the agents’ measurements?

Are there any papers out there that deal with this issue?