A collection of notes, projects, and essays.
Idle Frontier is a clicker game that challenges you to build a frontier language model in 1000 days! In this blog post, I'll talk about some reflections on developing a game with the help of large language models.
I was invited to give a talk at Analytics & AI Association of the Philippines (AAP) on FilBench, and in general, building Filipino LLMs. This blog post covers some of my thoughts on this topic.
A long reflection about my journey from industry to grad school applications—spanning multiple countries, jobs, and experiences. This is not an advise post, but I hope you'll find something valuable along the way.
This National Language Month, I'm proud to introduce FilBench, a big step forward in Filipino NLP evaluation. This work was also accepted at EMNLP Main! Read to learn more about this project.
Here is my field report from the ACL 2025 Conference in Vienna, Austria. Overall, it was a great experience: the vibes are good and I'm happy to have met the larger NLP community!
Just a fun weekend experiment on model-context protocol (MCP): I asked several tool-calling LLMs to draw a 4-frame spritesheet of a swordsman performing a slash attack using an Aseprite MCP I built. The results were interesting!
The rise of LLMs is forcing us to rethink Filipino NLP. But there's still a ton of work to do—just not the stuff you might think. Here's my take on what's worth doing, what's a waste of time, and where Filipino NLP research should be heading.
Last month, I had another guest lecture, this time in Dr. Charibeth Cheng's graduate class in DLSU. Here, I talked about the craft of building small-scale yet effective NLP models for Filipino in the face of today's large language models.
Can we spot differences between preference pairs just by looking at their word embeddings? In this blog post, I want to share my findings from examining lexical distances between chosen and rejected responses in preference datasets.
A few weeks ago, I held a guest lecture at University of North Carolina Charlotte on how we can use large language models for annotation in the context of argument mining and fact verification. Here are the contents of that lecture in blog post format.