Long-horizon brain signal generation with tokenizers + frontier VLMs
GitHub - ricsinaruto/brain-gen: Long-horizon brain signal generation with tokenizers + frontier VLMs
https://arxiv.org/pdf/2601.20138
A local, AI-powered text completion tool for web-based editors (e.g. gmail, notion)
https://github.com/ricsinaruto/shadow-clone-py
A proof-of-concept real-time silent speech BCI using EMG electrodes
https://github.com/ricsinaruto/emg-silent-speech-bci
.gif)
A wrapper around tensor2tensor to flexibly train, interact, and generate data for neural chatbots. Transformer-based dialog modeling back in 2017.
https://github.com/ricsinaruto/Seq2seqChatbots
https://arxiv.org/pdf/1908.08835
Easily simulate the change of dipole moments of complex protein structures and provide a 3D UI in OpenGL. There are also several searching algorithms implemented to find protein structures that satisfy user defined logic functions.
https://github.com/ricsinaruto/ProteinStructureSearch-Simulation
A lightweight repo for filtering dialog data with entropy-based methods
https://github.com/ricsinaruto/NeuralChatbots-DataFiltering
https://arxiv.org/pdf/1905.05471
A lightweight repo for automatic evaluation of dialog models using 17 metrics
https://github.com/ricsinaruto/dialog-eval
GPT-style MEG generation via quantization + channel / subject / condition embeddings
https://github.com/ricsinaruto/MEG-transfer-decoding
https://arxiv.org/pdf/2404.09256
Train Wavenet-based group-level models on MEG data, and uncover neuroscientifically interpretable information
https://github.com/ricsinaruto/MEG-group-decode
https://arxiv.org/pdf/2205.14102
MMG gesture recognition model: CNN–LSTM with custom data augmentation; >80% cross-participant accuracy (30 participants, 9 gestures)
Collected and analyzed 40 hours of inner-speech EEG/MEG/OPM data
An open-source pipeline for building dialogue datasets in many languages using books from Project Gutenberg
https://github.com/ricsinaruto/gutenberg-dialog
