AI Researcher (LLM) @Thoughtworks || PhD KAUST
Tiny-GRPO. minimal GRPO implementation from scratch
104Mini-LLaVA. A minimal implementation of LLaVA-style VLM with interleaved image & text & video processing ability.
99Evolutionary-Model-Merge. Unofficial Implementation of Evolutionary Model Merging
42CoT-Reasoning-without-Prompting. Unofficial Implementation of Chain-of-Thought Reasoning Without Prompting
35repo-viewer. Visualize any repo or codebase into diagram or animation
24selective-attention-transformer. Unofficial Implementation of Selective Attention Transformer
20STaR. Implementation of STaR: Bootstrapping Reasoning With Reasoning
4VocabularyCurriculum. Vocabulary curriculum improves scaling
4paper-read-2024-Jan-Apr. Paper reading run from 17. Jan 2024 - 17. Apr 2024
3CharacterLM. vocabulary curriculum + LLM
3RuleAlign. Rule-Based Adherance evaluation for LLM Agent
3llama3-sft. Supervised FineTuning with Llama-3
3auto-test. Automatic Chat with LLM & Test
2surfer. Voice Interface on Web Browsing
2Multilevel-Ensemble-Kalman-Bucy-Filter. This repository contains an implementation of the Multilevel Ensemble Kalman Bucy Filter. Code for HPC implementation with MPI cores, designed to run on Supercomputer Shaheen is also included. The algorithm deals with linear Gaussian filtering problems in continuous time and has appealing performance in high dimensions.
2RagAgent. Get some information locally with LLM
2Iterative_graph_alignment. Official Implementation of Iterative Graph Alignment https://arxiv.org/abs/2408.16667
2GeniusUpload. Why not ask genius mind instead of working on them ourselves?
2waveletGPT. Experimental code for wavelet GPT and Fourier series expression
2CS194-26-Computer-Vision-and-Computational-Photography. Jupyter Notebook
2diamond. DIAMOND (DIffusion As a Model Of eNvironment Dreams) is a reinforcement learning agent trained in a diffusion world model. NeurIPS 2024 Spotlight.
1GANs. Simple Implementation of implementations of Generative Adversarial Networks (GANs) using Tensorflow.
1concept-encoding. experiment with concept encoding in LLM
1abstraction-learning. Experiment for abstraction learning
1Normalizing-Flows. This repository contains examples of simple implementation of NF. Normalizing Flows are generative models which produce tractable distributions where both sampling and density evaluation can be efficient and exact.
1MiniAgent. Chinese ChatBot
1c3po. Jupyter Notebook
1MCVD. Video prediction with Diffusion Models
1AnnoAttribute. Annotation Tool for Decomposable Attributes
1FTchatter. Chat with FineTune model served with vLLM
1DenseNet. Densely Connected Convolutional Neural Network implementation in Pytorch
1vision-plus. Vision language model from scratch
1human-eval. Code for the paper "Evaluating Large Language Models Trained on Code"
1translate-thai. Translation for Thai Language
1llama3-function-call. training function-calling llama-3 agent with mlx
1CS231a-Computer-Vision-From-3D-Reconstruction-to-Recognition. Jupyter Notebook
1detr-object-detection. Jupyter Notebook
1PairRM-with-mps. PairRM but with MPS support
1adaptor. Adapt LLM with human-level efficiency
1CarND-Behavioral-Cloning-Project. Deep Learning for Self-Driving Cars
1payflow. Handling your document for you.
1Camera-Calibration. Jupyter Notebook
1Chicken-Planner. Plan as a chicken
1thought-compiler. Agent to help us in research
11bitNet. Experiments with 1Bit network
1process-supervision. Experiments with Procedure Supervision
1Paper-Read-April. Record of read paper in April
1align-llama3-with-reft. Fast & Efficient Alignment with Representation FineTuning & Llama3
1claude-agent. Claude Agent is built on Anthropic models to automate digital operation
1Alignment-experiment. Alignment Experiments
1fast_minbpe. Jupyter Notebook
1Transformer-MachineTranslation. Jupyter Notebook
1HPC_with_Python. Notebooks cataloging packages to enable using High Performance Computing on mulitcore and mulitnode architecture and clusters.
1Bayesian-Neural-Networks. Pytorch implementations of Bayes By Backprop, MC Dropout, SGLD, the Local Reparametrization Trick, KF-Laplace, SG-HMC and more
1Question-Answering-AI-with-BERT. Jupyter Notebook
1