PlayMaths
A playground where you can interact with widgets to solidify your mathematical understanding
Posts by Collection
engineeringProjects
Third Eye
A free online tool that aims to offer various computer vision services. Currently though, there is only one functionality available, object detection. Using our tool, you can detect objects from videos and photos in real time or asynchronously.
Burmese G2P
Myanmar (Burmese) 🇲🇲 Language Grapheme to Phoneme Converter for automatic speech recognition (ASR) and text-to-speech (TTS)
Space Invader
A classic arcade style 2D shooting game wherein the player spaceship shoots at the enemies and bosses while avoiding their attacks.
essays
Analyzing Moral Disengagement in Never Let Me Go and Nine Lives through Political and Economic Lens
When ethics clash with necessity, people create justifications to ease their guilt. Through the lens of human cloning, Never Let Me Go and Nine Lives expose how moral disengagement shapes society.
History of AI Development in China and its Entrepreneurial and Innovative Capabilities to Lead AI World Order
DeepSeek upset Silicon Valley’s assumptions about Chinese AI capabilities, but the signs were there all along. This piece traces how decades of entrepreneurial grit and strategic government planning quietly set the stage for China’s AI revolution, exploring some narratives in AI in the 20th century of China.
publications
WHEN TOM EATS KIMCHI: Evaluating Cultural Awareness of Multimodal Large Language Models in Cultural Mixture Contexts
Published in Workshop on Cross-Cultural Considerations in NLP (C3NLP) @ NAACL 2025, 2025
In a highly globalized world, it is important for multi-modal language models to correctly recognize visuals in mixed-cultural settings. This paper examines the robustness of MLLMs to mixed cultures by constructing MixCuBe, a cross-cultural awareness benchmark of images and evaluating SOTA MLLMs on it.
Physics-aware Multi-Object 3D Scene Reconstruction (in Progress)
Published in TBD, 2026
Recently, research in 3D reconstruction shifts from achieving consistency in mere appearance and geometry to attaining physically plausible models of the scene or the object. For this problem, while test-time optimization approaches takes hours to optimize reasonble physical parameters of even a single object, the generalizability of feed-forward approaches is too limited. We are currently striving to overcome the shortcomings of both approaches and provide a simulation pipeline that is easily generalizable.
researchProjects
Advanced Passage Retrieval
A NLP research into passage retrieval, the task of extracting top-k pertinent passages from a dataset as the output given a query as the input. We use the BM25 model as a benchmark to explore retrieval models that can achieve better accuracy specifically on the SQuAD1.1 dataset.
Dynamic Brain Connectome Learning
A set of graph machine learning architectures for learning temporal and spatial patterns of brain activation from fMRI images. Two downstream tasks are implemented: (1) brain activation prediction duration language tasks (link prediction) and (2) performance prediction from neural patterns (graph regression)
RenderFormer with Linear Attention
Renderformer is an end-to-end fully data-driven transformer-based rendering pipeline developed by Microsoft. Due to its usage of vinalla transformer architecture, it has time complexity O(N^2). We tried to achieve linear time complexity by using a linear attention mechanism, namely Performer (FAVOR++).
talks
SAP Analytics Cloud Training Permalink
Data analytics training using SAP analytics cloud to familiarize people around Myanmar with the theory of data science and hands-on experience with the tool. It also aims to promote ASEAN Data Science Explorer competition.
Overview of Programming Permalink
Introduction to the landscape of programming to the general public with emphasis on its relation with computer science and learning advice.
Hub&Hack 2022 Hackathon Python Training Permalink
A quick training on Python programming language for the participants of a national hackathon, named Hub&Hack, hosted by Thate Pan Hub.
Breaking into Artificial Intelligence Permalink
A taste of Artificial Intelligence to the general public
KAIST Undergraduate Admission Info & Tips Sharing Permalink
Info & tips sharing about KAIST Undergraduate Admission intended for students from Myanmar
teaching
AI for Teenagers Permalink
An introductory AI course for teenagers and any novice of all ages alike. It starts off by exploring some subfields of AI before deep diving into machine learning, where you will learn three major types of learning and some of their algorithms. After that, it will give students a sense of the working of artificial neural networks and exposure to some phenomena of AI in society.