yinyajun

EasyNLP

EasyNLP: A Comprehensive and Easy-to-use NLP Toolkit

simple-evals

trian_ppo

LLM

Chinese-LLaMA-Alpaca-3

中文羊驼大模型三期项目 (Chinese Llama-3 LLMs) developed from Meta Llama 3

RL4LMs

A modular RL library to fine-tune language models to human preferences

Awesome-Text2SQL

Curated tutorials and resources for Large Language Models, Text2SQL, Text2DSL、Text2API、Text2Vis and more.

ten_framework

TEN, the Next-Gen AI-Agent Framework, the world's first truly real-time multimodal AI agent framework.

Hands-on-RL

https://hrl.boyuai.com/

DWPose

"Effective Whole-body Pose Estimation with Two-stages Distillation" (ICCV 2023, CV4Metaverse Workshop)

mmaction2

OpenMMLab's Next Generation Video Understanding Toolbox and Benchmark

Kolors

Kolors Team

agents

Build real-time multimodal AI applications 🤖🎙️📹

I2V-Adapter

I2V-Adapter: A General Image-to-Video Adapter for Diffusion Models

vast-pyworker

LongLoRA

Code and documents of LongLoRA and LongAlpaca (ICLR 2024 Oral)

self-instruct

Aligning pretrained language models with instruction data generated by themselves.

Awesome-LLM-Inference

📖A curated list of Awesome LLM Inference Paper with codes, TensorRT-LLM, vLLM, streaming-llm, AWQ, SmoothQuant, WINT8/4, Continuous Batching, FlashAttention, PagedAttention etc.

Awesome-Efficient-LLM

A curated list for Efficient Large Language Models

litgpt

20+ high-performance LLMs with recipes to pretrain, finetune and deploy at scale.

lit-llama

Implementation of the LLaMA language model based on nanoGPT. Supports flash attention, Int8 and GPTQ 4bit quantization, LoRA and LLaMA-Adapter fine-tuning, pre-training. Apache 2.0-licensed.

MANIQA

[CVPRW 2022] MANIQA: Multi-dimension Attention Network for No-Reference Image Quality Assessment

minSDXL

Huggingface-compatible SDXL Unet implementation that is readily hackable

Awesome-Backbones

Integrate deep learning models for image classification | Backbone learning/comparison/magic modification project

examples

TensorFlow examples

CValues

面向中文大模型价值观的评估与对齐研究

t-zero

Reproduce results and replicate training fo T0 (Multitask Prompted Training Enables Zero-Shot Task Generalization)

architecture-objective

attention-rank-collapse

[ICML 2021 Oral] We show pure attention suffers rank collapse, and how different mechanisms combat it.

svd-temporal-controlnet