👋 Hi, nice to meet you! I’m Emmanuel G. Maminta.

• 🎓 Artificial Intelligence Ph.D. student (ongoing; expected 2027)
• 🏫 University of the Philippines
• 📍 Metro Manila, Philippines
• 📧 egmaminta [at] up.edu.ph

Short Bio

I am a fourth-year Artificial Intelligence Ph.D. student at the Ubiquitous Computing Laboratory (UCL) in the University of the Philippines, advised by Prof. Rowel O. Atienza.

I obtained my Bachelor of Science in Electrical Engineering (B.S. EE) with honors in 2022, and my Master of Engineering in Artificial Intelligence (M.Eng. AI) in 2025, both from the University of the Philippines. I have gained more than half a decade of experience in research and development, software engineering, and project management, with a focus on the field of AI and Machine Learning (ML).

My primary research interests are the following:

• Embodied AI agents or agentic AI, and their applications in the retail industry (Retail Intelligence)
• Multimodal deep learning (vision & language) including fusion techniques, foundation models (FMs), and vision language models (VLMs)
• Pre-training methods and how to efficiently utilize single GPU resource for such methods
• Reinforcement learning-based post-training methods in large language models (LLMs), specifically reward modeling, preference learning, and policy optimization algorithms
• Computer vision including out-of-vocabulary (or in-the-wild) object detection and image segmentation

Other research interests include: AI safety and alignment, Chain-of-Thought (CoT) monitoring, prompt optimization.

---

News

• [Summer 2025] Contributed a customized adapter (AnyMaths) in the GEPA (Genetic-Pareto) framework for optimizing prompts and other text components specifically in word problems (mathematics) and LLM reasoning. [code:gepa] [paper]
• [Summer 2025] Published GEPA-Lite, a lightweight version of GEPA under the MIT License via GitHub. Project is starred and endorsed by one of the paper’s original authors, Lakshya A. Agrawal. [gepa-lite]
• [Summer 2025] One paper is accepted in ICCV 2025 Workshop on RetailVision. [arXiv] [webpage] [paper]
• [Summer 2025] Presented research findings in a closed-door technical workshop with Analog Devices (ADI) Philippines. [arXiv] [webpage] [paper]
• [Summer 2025] Completed all coursework for M.Eng. AI program. [capstone project]
• [Summer 2024] Presented research findings (with Nikko Carlo A. Yabut and SRPH) in the 1st National Electrical, Electronics, and Computer Engineering Conference (NEECECon). [abstract]
• [Spring 2023] Affiliated with UCL, rostered as one of the graduate researchers for a collaborative project with Samsung Research & Development Institute Philippines (SRPH) about out-of-vocabulary object detection in VLMs.
• [Autumn 2022] Co-taught EEE 111 Introduction to Programming and Computation and EEE 121 Data Structures and Algorithms with Prof. Jhoanna Rhodette I. Pedrasa at the Electrical and Electronics Engineering Institute (EEEI) of the University of the Philippines.
• [Autumn 2022] Began M.Eng-Ph.D. studies as part of the pioneering “Turing Batch” in the Philippines’ first AI graduate program.