AlphaLab: Autonomous Multi-Agent Research with Advanced LLMs
Discover AlphaLab, an autonomous multi-agent system using frontier LLMs to optimize GPU kernels, LLM pretraining, and traffic forecasting efficiently.
When Should AI Act or Escalate? Insights on Automation
Explore how language models decide between acting or escalating tasks in automation, improving AI decision-making under uncertainty across key sectors.
FluidFlow: Generative Surrogate Model for CFD on Meshes
Discover FluidFlow, a flow-matching generative model offering accurate fluid dynamics surrogates on unstructured meshes without preprocessing.
QCFuse: Efficient Query-Centric Cache Fusion for RAG
Discover QCFuse, a query-centric cache fusion system that boosts RAG inference efficiency by 40% while maintaining accuracy in large language models.
CSAttention: Fast, Accurate Sparse Attention for LLMs
Discover CSAttention, a novel sparse attention method boosting LLM inference speed by 4.6x while maintaining accuracy with high sparsity and long contexts.