Search papers, labs, and topics across Lattice.
University of Leicester
2
15
5
8
A single verification prompt is insufficient; learning a reliability score from multiple, complementary perspectives unlocks robust self-verification for VLMs.
Gradient spikes in LLM training can be 1000x larger than normal, but a new optimizer, SPAM, tames them with momentum reset and spike-aware clipping, boosting performance and memory efficiency.
