Search papers, labs, and topics across Lattice.
Affiliation:
2
0
4
Detecting a backdoor trigger does not mean you can stop it: SAE features in attention and MLPs reliably flag trojan inputs but lack causal control, whereas sparse directions in the residual stream hold full steering power over backdoor execution.
Machine translation benchmarks have functionally saturated, but pairing human-authored failure cases with deterministic verification rules reveals critical multimodal blind spots that automated metrics consistently miss.
