Search papers, labs, and topics across Lattice.
6
0
9
Interference-aware service placements can drastically enhance response performance in cloud-native applications by explicitly controlling for cross-application resource competition.
Modularity in neural architectures can dramatically improve compositional learning, but only in lower-dimensional settings where task representations are rich and nuanced.
Rovers can now navigate more intelligently by querying real-time semantic maps with natural language, thanks to a novel confidence-aware mapping approach.
Observational metrics fail to predict causal expert importance in Mixture-of-Experts models, undermining common pruning practices.
LLMs can synthesize formal safety rules from natural language goals, offering a path to more robust and verifiable AI systems in safety-critical domains.
LLM-generated debugging explanations are often vague or misleading, but this work shows you can make them dramatically better by carefully curating the context provided to the LLM.
