BioVLM

BioVLM

Evidence-enriched data synthesis pipeline from biomedical papers, training BioVLM-8B that surpasses GPT-5.2 on LAB-Bench.

LLM Biomedical VLM
LLM-Monitor

LLM-Monitor

Low-cost, evolvable privacy violation monitoring system for LLMs, 15x faster than GPT-5.2.

LLM Privacy Safety
AI4Math Sketchpad

AI4Math Sketchpad

Grant-funded system for converting mathematical proofs into structured data for auto-formalization.

AI Mathematics Formal Verification
BTMR-Paper

BTMR-Paper

Insanely Fast Paper Reading Tool - An AI-powered web application for extracting, analyzing, and summarizing academic papers

AI LLM Research Python React
Bili-Investigate

Bili-Investigate

A Streamlit-based web application for tracking Bilibili content creators' video updates with smart incremental fetching

Python Web Application Streamlit Data Collection
YA-PapersWithCode

YA-PapersWithCode

Yet Another Papers With Code - A modern recreation of the Papers With Code platform with AI-powered semantic search

AI LLM Research Python React TypeScript
ContextKeeper

ContextKeeper

AI assistant for RTX GPU users with extensible plugin ecosystem - A community fork of NVIDIA G-Assist

AI Python C++ Plugin System Voice Assistant
Autoreader

Autoreader

Semantic searching in daily arXiv papers.

LLM VectorDB
Nerif

Nerif

LLM-powered Python, make you write Python code with natural language.

LLM
TREVA

TREVA

Never Miss Your Lecture Again. From the 2023 HacktheBurgh Hackathon, this project aims to convert lecture recordings into understandable and readable notes.

Deep Learning
DeepReader

DeepReader

DeepReader is a tool that allows you to discuss books with AI deeply.

LLM Multi-agent
AI QR Code

AI QR Code

Stable Diffusion + QR Code Generation. Generate a combined QR code for couples.

LLM Stable Diffusion
Find-A-Tenant

Find-A-Tenant

AI rental agent, helps you find a house based on fuzzy needs. With AI Fuzzy Search Agent, Automatic Metadata Extraction Agent, and more.

LLM VectorDB
QED Simulation on GPU

QED Simulation on GPU

High-performance Monte-Carlo QED simulation with 1500x-5000x speedup, recognized by Nobel Laureate Frank Wilczek.

GPU CUDA Physics HPC