aiforscience

Litex: The Language Where Mathematics Verifies Itself.

Generative Flow Networks - GFlowNet

ScienceBuddy: Recursive-in-Recursive Self-Improvement for Interactive Scientific Agents

Open-source AI infrastructure for materials science

Derivative-Free Guidance in Diffusion Models with Soft Value-Based Decoding. For controlled generation in DNA, RNA, proteins, molecules (+ images)

Discovery Foundation Models (DFMs): From solving tasks to discovering new knowledge.

FourierFlow: Frequency-aware Flow Matching for Generative Turbulence Modeling

AI驱动的分波分析自动化。

List of job market in AI4S

A comprehensive system for evaluating AI research papers using advanced language models with asynchronous processing and concurrent evaluation capabilities.

This project is an open-source project based on my AI algorithm, which won second place nationwide in the application track of AI algorithms in unknown phases of new materials at the 7th Global Algorithm Elite Challenge. It is used for the prediction of unit cell parameters.

AI for Materials Science Training and Development

AI agent specialised in study, learning, and scientific research and discovery.

AI-driven brain network analysis for autism research using the ABIDE dataset. This project builds a pipeline from resting-state fMRI ROI time series to GNN classification, interpretability, literature grounding, and in-silico intervention.

An open computational research project on lymphangioleiomyomatosis (LAM) based on public biomedical data.

Multi-agent research testing whether CMIP6 misses recent tropical rainfall because mean-state biases mute the circulation response to land-sea warming, not because the thermal drivers are missing.