Search papers, labs, and topics across Lattice.
Reasoning-Lab/SPUR BUPT-Reasoning-Lab, Beijing University of Posts and Telecommunications
4
0
6
Current vision-language models are surprisingly bad at interpreting scientific figures, failing to match expert-level reasoning on a new benchmark of experimental images.
Optimizing treatments for time-to-event outcomes just got faster: bandit algorithms can now learn near-optimal survival analysis policies online.
LLMs' hallucinations stem from a "gray zone" of internal belief ambiguity near knowledge boundaries, and geometric denoising in the latent space offers a surprisingly effective way to purge it.
Current memory systems, despite their complexity, are surprisingly worse than naive RAG when applied to continuous lifelogging scenarios, revealing a critical need for better context preservation.
