Neuron-Based Rule Extraction for Explainable Large Language Models
Discover MechaRule, a novel neuron-anchored method that improves rule extraction and interpretability in large language models for explainable AI.
ARIS: AI-Driven Autonomous Research with Multi-Agent Collaboration
Discover ARIS, an AI framework enabling autonomous research through adversarial multi-agent collaboration for validated and reliable scientific outputs.
Efficient On-Device Bipolar Agitation Detection with MP-IB
Discover MP-IB, a mixed-precision AI framework for real-time bipolar agitation detection using voice biomarkers on low-power devices.
Structured Diffusion Bridges for Flexible Modality Translation
Discover how structured diffusion bridges enable high-quality modality translation with flexible supervision, reducing reliance on fully paired datasets.
Enhancing Multilingual AI Safety with Self-Distillation
Discover how self-distillation improves safety alignment in multilingual AI, boosting protection in low-resource languages without extensive data.