Search papers, labs, and topics across Lattice.
The Pennsylvania State University State College
2
0
6
LLMs can learn to abstain from answering questions they're unsure about with state-of-the-art accuracy by dynamically re-weighting abstention rewards based on trajectory consistency during training.
VLMs can be easily tricked into "hallucinating" object relationships with simple image rotations or noise, revealing a surprising fragility in their multimodal reasoning.
