RadLite: Efficient CPU Radiology AI with LoRA Fine-Tuning
RadLite uses LoRA fine-tuning on small language models for multi-task radiology AI, enabling CPU deployment and high accuracy in clinical settings.
Unifying Decision Trees and Diffusion Models for AI
Discover how unifying decision trees with diffusion models boosts AI efficiency, interpretability, and performance in machine learning.
Agent Capsules: Optimize Multi-Agent LLM Pipelines Efficiently
Discover how Agent Capsules improve multi-agent LLM pipelines by reducing tokens and enhancing output quality with adaptive execution strategies.
Scalable Learning in Recurrent Spiking Neural Networks
Discover scalable learning in structured recurrent spiking neural networks using local plasticity without backpropagation for efficient AI.
Mitigating Social Bias in LLM-Generated Code Effectively
Explore how the Fairness Monitor Agent reduces social bias by 65% in LLM-generated code, improving fairness and functional correctness across 343 tasks.