Methods, systems and one piece of hardware. Each entry links to the paper, the code, or a demo you can run.
We reveal that transformers trained autoregressively naturally encode causal structures — gradient attributions directly recover underlying causal graphs without any explicit causal objectives.
A reinforcement learning framework that generalizes to unseen environments by learning language-controlled causal components and recombining them — with identifiability guarantees.
An LLM-powered agent that runs the whole causal analysis loop — diagnosing the data, selecting and configuring the right method from 20+ options, checking its own results, and producing an inspectable report.
Instead of designing a causal discovery algorithm, we learn one — from a microprocessor whose every causal edge can be established by intervention. It outperforms human-designed methods on silicon, simulated fMRI and gene networks.
A millisecond-level phase locked neural feedback system based on OpenBCI for real-time alpha wave regulation, integrating acquisition, phase estimation and stimulation on one chip.
A robust deep learning pipeline for segmenting neuronal cells in microscopy images — Cascade Mask R-CNN X152 with semi-supervised pseudo-labelling and cascade IoU fusion.
Paper
arXiv
Code
Demo
Video
Paper
Write-up