Skip to content

Latest commit

 

History

History
188 lines (164 loc) · 27.4 KB

README.md

File metadata and controls

188 lines (164 loc) · 27.4 KB
Raw

Awesome Llama 2

Llama 2 is a second-generation open-source large language model (LLM) from Meta. It can be used to build chatbots like ChatGPT or Google Bard.

Table of Contents

GitHub projects

Articles & Blogs

Online Courses

Books

Research Papers

  • Llama 2: Open Foundation and Fine-Tuned Chat Models: This paper presents Llama 2, a collection of pretrained and fine-tuned large language models optimized for dialogue use cases. The models outperform open-source chat models on various benchmarks and provide detailed insights into fine-tuning and safety improvements.
  • Effective Long-Context Scaling of Foundation Models: This paper introduces a series of long-context large language models (LLMs) built by continual pretraining from Llama 2. These models support effective context windows of up to 32,768 tokens and achieve consistent improvements on regular and long-context tasks over Llama 2.
  • Code Llama: Open Foundation Models for Code: The authors release Code Llama, a family of large language models for code based on Llama 2. These models provide state-of-the-art performance among open models, infilling capabilities, support for large input contexts, and zero-shot instruction following ability for programming tasks.
  • Financial News Analytics Using Fine-Tuned Llama 2 GPT Model: This paper explores the fine-tuning of Llama 2 GPT large language model for multitask analysis of financial news. The study demonstrates the model's ability to analyze, highlight main points, summarize, and extract named entities with appropriate sentiments, enabling structured financial news analysis.
  • Efficient and Effective Text Encoding for Chinese LLaMA and Alpaca: The paper proposes a method to augment LLaMA with capabilities for understanding and generating Chinese text. It extends LLaMA's existing vocabulary with Chinese tokens, incorporates secondary pre-training using Chinese data, and fine-tunes the model with Chinese instruction datasets, enhancing its encoding efficiency and semantic understanding of Chinese.
  • LLaMA: Open and Efficient Foundation Language Models: This paper introduces LLaMA, a collection of foundation language models ranging from 7B to 65B parameters. These models are trained on publicly available datasets exclusively, without resorting to proprietary datasets, and outperform GPT-3 on various benchmarks.
  • Fine-Tuning Llama 2 Large Language Models for Detecting Online...: The paper proposes an approach to detecting online sexual predatory chats and abusive language using the open-source pretrained Llama 2 7B-parameter model. The approach is generic, automated, and fine-tunes the model using datasets in different languages.
  • Comparing Llama-2 and GPT-3 LLMs for HPC kernels generation: This paper evaluates the use of the open-source Llama-2 model for generating high-performance computing kernels on different parallel programming models and languages, comparing its accuracy to GPT-3.
  • LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model: This paper presents LLaMA-Adapter V2, a parameter-efficient visual instruction model that improves the ability of large language models to handle open-ended visual instructions.
  • ChatDoctor: A Medical Chat Model Fine-Tuned on a Specialized LLaMA: This research focuses on enhancing the medical knowledge of large language models, like ChatGPT, by creating a specialized language model with improved accuracy in medical advice.
  • Label Supervised LLaMA Finetuning: This paper explores the potential of leveraging latent representations from LLaMA large language models for supervised label prediction, particularly in sequence and token classification tasks.
  • Multi-Task Instruction Tuning of LLaMa for Specific Scenarios: This work investigates the potential of using LLaMA for targeted scenarios, focusing on the writing assistance scenario, and demonstrates the improvement achieved through fine-tuning LLaMA on writing instruction data.
  • PMC-LLaMA: Towards Building Open-source Language Model for Medical Applications: This paper describes the process of adapting a general-purpose language model towards medical domain applications, including data-centric knowledge injection and comprehensive fine-tuning for alignment with domain-specific instructions.
  • Llama 2: Open Foundation and Fine-Tuned Chat Models: This work develops and releases Llama 2, a collection of pretrained and fine-tuned large language models optimized for dialogue use cases, which can be a suitable substitute for closed-source models.
  • LoRA Fine-tuning Efficiently Undoes Safety Training in Llama 2: This paper explores the robustness of safety training in language models by subversively fine-tuning the public weights of Llama 2-Chat models, using the low-rank adaptation (LoRA) technique.

Videos

Tools & Software

Conferences & Events

Slides & Presentations

Podcasts

This initial version of the Awesome List was generated with the help of the Awesome List Generator. It's an open-source Python package that uses the power of GPT models to automatically curate and generate starting points for resource lists related to a specific topic.