Advanced Computing (Supercomputing; Edge computing; Cloud computing and services; High-performance data storage and data centers; Computing architectures; Advanced modeling and simulation; Data processing and analysis techniques), The Science and Engineering of Evaluation (Evaluatology and Benchmarkology), Artificial Intelligence (AI system frameworks, AI assurance and assessment techniques; Generative AI systems, multimodal and large language models; Synthetic data approaches for training, tuning, and testing; Technologies for improving AI safety, trust, security, and responsible use), and Operating Systems.

Evaluation is one of the fundamental activities of humanity. Advanced computing systems will help unleash human creativity and create a better world. At the same time, building a positive and non-destructive life community (including humans, various animals, and plants) contributes to the enhancement of the dignity and value of life. My three future research goals include: (1) Establishing a unified discipline of Evaluation Science and Engineering; (2) Researching foundational theories, models, and methods to construct and manage simple, secure, and fair advanced computing-based social infrastructures; (3) Investigating the fundamental institutions and major technical barriers of a positive and non-destructive life community (including humans, animals, and plants).

In the next five years (2025-2030), I plan to engage in the following research topics:

• Evaluatology and Benchmarkology. I have been fortunate enough to coin these terms (with contributions from my doctoral student, Hongxiao Li to the term of Evaluatology). Evaluation is a fundamental issue across all disciplines. Unfortunately, evaluation methods in most disciplines are often ad-hoc and empirical. Developing unified principles and methods for evaluation is a challenging problem. I have proposed the fundamental concepts, unified terminology, axioms, theories, methodologies, and foundational issues of Evaluatology. I will further advance Evaluatology and Benchmarkology in disciplines such as safety-critical, mission-critical and business-critical AI systems, intelligence, energy efficiency, science and technology, social infrastructure, computers, medicine, and pharmaceuticals and safety (especially AI safety). I plan to publish five volumes English monograph on Evaluatology and Benchmarkology.
• Research on advanced computing-based social infrastructures. I will introduce new abstract simplifications for advanced computing-based social infrastructures, ensuring their security and fairness based on Evaluatology. I will further refine the Open Computer Systems Initiative (OpenCS) that I proposed earlier. The OpenCS initiative will facilitate research on advanced computing-based social infrastructures, without reinventing the wheel. Instead, I will propose new abstractions and methods to address these significant challenges.
• Research on the fundamental institutions and major technical barriers of a constructive rather than destructive life community. I will systematically expound on this relatively utopian research objective and vision in the near future.