Search papers, labs, and topics across Lattice.
Normal University, NeoteAI
6
0
9
RadioDecomp is proposed, which treats a prior-guided predictor as a correctable base and uses deterministic residual refinement to learn its remaining predictable discrepancy, and is instantiate as RadioLSR (LoS-Shadow-Residual).
RACE reveals that a scalable statistical approach can vastly improve our understanding of neuron behavior in LLMs while slashing computational costs.
StateTrace reveals that enhancing memory structures in VLMs can drastically improve reasoning about invisible objects, achieving a remarkable 24.6-point performance increase on hidden-state tasks.
Scaling AEB systems with massive unlabeled data can lead to a 35% increase in accident-free driving mileage while maintaining a positive-to-false activation ratio over 100:1.
