This SCBX Group framework outlines 20 key principles for building smarter agentic AI systems, focusing on evaluation, context engineering, prompt design, and tool integration to improve how AI thinks and acts.

This Research by SCBX Group and Partners presents Typhoon-S, a framework for minimal open post-training designed to customize base models into highly capable regional assistants. It outlines requirements for general adoptability and sovereign capability to enable localized AI deployment with transparency.

This Research by SCBX Group and Partners introduces AudioJudge, a speech evaluation framework utilizing large audio models (LAMs) as judges. The system demonstrates a high correlation with human judgment across multiple audio attributes, offering a unified evaluation solution.

This Research by SCBX Group and Partners shows that Program-of-Thought (PoT) improves multilingual reasoning in LLMs by separating reasoning from execution. After fine-tuning, PoT outperforms Chain-of-Thought (CoT), and better reasoning leads to more accurate answers.