-
University of Seoul; LBox
- South Korea
- https://scholar.google.com/citations?user=M13_WdcAAAAJ&hl=en
Highlights
- Pro
Stars
š AI orchestration framework to build customizable, production-ready LLM applications. Connect components (models, vector DBs, file converters) to pipelines or agents that can interact with your daā¦
Generative Agents: Interactive Simulacra of Human Behavior
High-quality datasets, tools, and concepts for LLM fine-tuning.
Course to get into Large Language Models (LLMs) with roadmaps and Colab notebooks.
DevOps Roadmap for 2024. with learning resources
A fast implementation of T5/UL2 in PyTorch using Flash Attention
š„ Transform PDF to JSON or Markdown with ease and speed š£
Python implementations (on jupyter notebook) of algorithms described in the book "PRML"
Implementation of Alphafold 3 in Pytorch
DSPy: The framework for programmingānot promptingāfoundation models
Implementing a ChatGPT-like LLM in PyTorch from scratch, step by step
Distilabel is a framework for synthetic data and AI feedback for engineers who need fast, reliable and scalable pipelines based on verified research papers.
A repository for research on medium sized language models.
Retrieval and Retrieval-augmented LLMs
A Protein Large Language Model for Multi-Task Protein Language Processing
Running large language models on a single GPU for throughput-oriented scenarios.
Complex-based Ligand-Binding Proteins Redesign by Equivariant Diffusion-based Generative Models
Code for the paper "Bottleneck Minimal Indexing for Generative Document Retrieval" accepted by ICML 2024
The largest collection of PyTorch image encoders / backbones. Including train, eval, inference, export scripts, and pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (Vā¦
Code repo for CLERC: A Legal Precedent Dataset for Case Retrieval and Retrieval-Augmented Analysis Generation
CoLLaM: A Comprehensive Benchmark for Evaluating Large Language Models in Legal Domain
ā”FlashRAG: A Python Toolkit for Efficient RAG Research
Freeing data processing from scripting madness by providing a set of platform-agnostic customizable pipeline processing blocks.