Search papers, labs, and topics across Lattice.
5
0
7
LLM-translated code can be slower than human-written code, but SwiftTrans bridges this gap with a novel two-stage framework that boosts both correctness and efficiency.
Reducing reasoning tokens while boosting accuracy, SuCo transforms how LRMs approach problem-solving by focusing on sufficiency rather than excess.
Achieving over 5% improvement in coreference resolution without the need for resource-intensive training on target datasets could revolutionize multimodal AI applications.
MLLMs excel at precision but falter dramatically in extracting complete product specifications from multiple images, with a mere 49.9% recovery rate.
Transforming a single image into a fully navigable 3D world in real-time could revolutionize how we interact with visual environments.
