Search papers, labs, and topics across Lattice.
4
0
7
5
Popular facts are harder to forget, but AdaPop can effectively unlearn them while retaining essential knowledge, outperforming existing methods by a significant margin.
VLA models retain basic knowledge but falter on complex semantics, revealing critical gaps in their adaptation from VLMs.
LLMs often fail to update their final predictions after interventions on intermediate reasoning steps, suggesting that these structures function more as influential context than stable causal mediators.
LLMs can fact-check surprisingly well without external retrieval, and the key is tapping into their internal representations, not just the output logits.
