Preventing Policy Violations in LLM-Based AI Agents
Explore how PhantomPolicy and Sentinel frameworks help detect and prevent hidden policy violations in LLM-based agents for better compliance.
Assessing Relational Reasoning in LLMs with REL Benchmark
Discover how the REL benchmark evaluates relational reasoning in large language models, revealing key insights on their performance with complex relations.
EMBER: Autonomous Cognitive AI with Spiking Neural Networks
Discover EMBER, a hybrid AI combining spiking neural networks and LLMs for autonomous cognitive behavior and rapid learning in advanced architectures.
AI System Enhances Thoracic Tumor Board Case Summaries
Discover how AI-driven chart summarization improves efficiency and accuracy in thoracic tumor board patient case reviews at Stanford.
Opinion-Aware Retrieval-Augmented Generation for Diverse AI
Discover how opinion-aware retrieval-augmented generation improves AI by embracing diverse viewpoints and enhancing sentiment and demographic representatio...