Author

Santos

h-index 1 View on Semantic Scholar

We have 1 of 11 papers

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Advanced Architectures for AI Self-Training:State-of-the-Art Methodologies and Performance Simulation

A robust framework that integrates RLAIF (Reinforcement Learning from AI Feedback) and a novel Multi-Agent Consensus Protocol (MACP) is proposed, demonstrating a 14.5% improvement in F1-score across highly complex multimodal domains compared to classical baselines.

Dheiver Francisco, Santos · 0 citations