Search papers, labs, and topics across Lattice.
1
0
2
1
PFNs, despite their strong empirical performance in causal inference, can suffer from prior-induced confounding bias, but a simple one-step posterior correction can restore frequentist consistency.
