Cost-Effective Network Topologies for MoE LLM Serving
Discover cost-effective network topologies that optimize Mixture-of-Experts LLM serving, improving performance while reducing infrastructure expenses.
MAEPose: Self-Supervised mmWave Human Pose Estimation
Discover MAEPose, a self-supervised model using mmWave radar for accurate, privacy-preserving human pose estimation with spatiotemporal learning.
Attention Redistribution Attack Threatens LLM Safety
Discover how the Attention Redistribution Attack bypasses safety in large language models, exposing vulnerabilities in AI attention mechanisms.
Why LLMs Fail in Strategic Play: Key Decision Gaps
Explore why LLMs struggle with strategic play due to gaps between observations, beliefs, and actions, impacting decision-making in complex games.
RSAT: Boosting Small Language Models for Accurate Table Reasoning
Discover how RSAT improves faithfulness and transparency in small language models for reliable table reasoning with structured attribution.