CUAD (NeurIPS 2021)

Extracting spatial and temporal world models from LLMs

A library for mechanistic interpretability of GPT-style language models

💠 An Obsidian plugin to create a table of contents in a note, that updates itself when the note changes

[ICME 2024] Official Repository for The Paper, PianoBART: Symbolic Piano Music Understanding and Generating with Large-Scale Pre-Training

TRAM: Benchmarking Temporal Reasoning for Large Language Models (Findings of ACL 2024)

NR/TSR

ZuCo Reading Task Classification Benchmark using EEG and Eye-Tracking Data

The implementation of "Systematic Analysis of Music Representations from BERT"

Model loading code for CochDNN auditory models (auditory models with cochleagram front end)

A game theoretic approach to explain the output of any machine learning model.

A PyTorch implementation of EfficientNet

Implementation of recent Deep Learning papers

🧑‍🏫 60+ Implementations/tutorials of deep learning papers with side-by-side notes 📝; including transformers (original, xl, switch, feedback, vit, ...), optimizers (adam, adabelief, sophia, ...), ga…

Implementation of FuseMoE for FlexiModal Fusion, NeurIPS'24

PySimpleGUI based DESKTOP APP to AUTO GENERATE SUBTITLE FILE (using free Google Speech Recognition API) and TRANSLATED SUBTITLE FILE (using unofficial online Google Translate API) for any video or …

Command line utility for forced alignment using Kaldi

WhisperX: Automatic Speech Recognition with Word-level Timestamps (& Diarization)

Reproduce CKA: Similarity of Neural Network Representations Revisited

Parameters for the neural network used in Kell, Yamins, Shook, Norman-Haignere, and McDermott. "A task-optimized neural network replicates human auditory behavior, predicts brain responses, and rev…

Chinese version of GPT2 training code, using BERT tokenizer.

MiniCPM-V 4.5: A GPT-4o Level MLLM for Single Image, Multi Image and High-FPS Video Understanding on Your Phone

AlwaysReddy is a LLM voice assistant that is always just a hotkey away.

This is the Personality Core for GLaDOS, the first steps towards a real-life implementation of the AI from the Portal series by Valve.