Longitudinal AI Health Agent Framework for Patient Support
Explore a multi-layer AI framework enabling personalized, coherent, and adaptive health support over time for improved patient engagement and decision-maki...
Geometric Evidence of Agent Identity in LLM Activation Space
Discover how agent identity acts as an attractor in LLM activation space, revealing persistent cognitive architecture across models like Llama and Gemma.
Memory Worth: Dynamic Memory Governance for AI Agents
Discover Memory Worth, a lightweight metric for AI agents to dynamically govern memory quality and improve task success through adaptive memory management.
Diagnosing Failures in Long-Horizon AI Agentic Systems
Explore how the HORIZON benchmark diagnoses failures in AI agents during complex long-horizon tasks, improving reliability and performance.
ArcDeck: Narrative Paper-to-Slide Generation Framework
ArcDeck uses multi-agent narrative reconstruction to convert academic papers into coherent slides, improving flow and clarity in presentations.