Search papers, labs, and topics across Lattice.
Address:
7
0
12
4
Premature fitness evaluations can bias evolutionary outcomes, but AdaControl reveals a path to more diverse and efficient robot designs.
VAKE reveals that over 80% of the knowledge activated through explicit priming is crucial for answering questions, showcasing a new pathway to enhance LLM factual accuracy.
Traditional trajectory-based training collapses policy diversity, but DART-SD's topology-aware approach ensures agents can explore optimal paths without penalty.
Task-specific optimization in multimodal models can significantly boost performance by dynamically reallocating shared computation resources between vision and language components.
A cooperative dual-altitude navigation strategy can improve aerial target localization success rates by over 30% while using significantly less adaptation data.
Unlock the power of cutting-edge photon-counting CT imaging on your existing routine chest CT scans, boosting lesion detection by 10-15%.
Robot control systems are shockingly vulnerable: JailWAM achieves an 84.2% success rate in jailbreaking state-of-the-art World Action Models to perform unsafe physical actions.
