I'm Arthur Jiang, a software engineer and Linux enthusiast currently studying as a sophomore at Emerald High School.
I have a deep passion for building tools that automate the mundane and exploring the depths of operating systems. When I'm not conducting research on LLMs or building window managers in the browser, you can find me customizing my Arch Linux setup or making boba.
Brewing joy, one boba tea at a time. Handling customer service and drink preparation in a fast-paced environment.
Conducted research on Bayesian reasoning performance in Large Language Models (LLMs) in the context of bandit simulation problems.
Currently a 10th-grade student focusing on Computer Science and Mathematics.
An AI-backed automation framework for completing repetitive tasks like web scraping and computer actions. 🏆 3rd Place at ModelMinds Hackathon
AI-powered calendar tool that automatically schedules events and reminders based on images captured by the user. 🏆 1st Place at EHS Hackathon
A vanilla JavaScript library for creating a full desktop-like OS experience entirely within the browser window.
An attempt at reverse engineering Rabbit's LAM (Large Action Model). The predecessor to Agent SURF.
A simple OS-like interface for the browser, focused on bypassing internet censorship in schools. Acquired in 2023.
This research paper evaluates probabilistic reasoning capabilities in Large Language Models through language-only decision tasks, specifically focusing on bandit simulation problems. The work explores how LLMs perform Bayesian reasoning in contextual scenarios without numerical computation.
View on OpenReview