Generative AI Literature

Large Language Model

• LLaMA: Open and Efficient Foundation Language Models (Feb 2023)  Link  PDF
• Stanford Alpaca: An Instruction-following LLaMA Model (March 2023)  LinkGithub
• Llama 2: Open Foundation and Fine-Tuned Chat Models (Jul 2023)  Link  PDF
• The RefinedWeb Dataset for Falcon LLM: Outperforming Curated Corpora with Web Data, and Web Data Only (Jun 2023)  Link  PDF
• Mistral 7B (Oct 2023)  Link  PDF

RAG

• Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks (Apr 2021)  Link  PDF
• Fine-tune the Entire RAG Architecture (including DPR retriever) for Question-Answering (Jun 2021)  Link  PDF
• Benchmarking Large Language Models in Retrieval-Augmented Generation (Sep 2023)  Link  PDF

Fine Tuning

• Fine-tuning language models to find agreement among humans with diverse preferences (Nov 2022)  Link  PDF
• Instruction Tuning for Large Language Models: A Survey (Oct 2023)  Link  PDF
• Self-Instruct: Aligning Language Models with Self-Generated Instructions (May 2023)  Link  PDF
• Synthetic Prompting: Generating Chain-of-Thought Demonstrations for Large Language Models (Feb 2023)  Link  PDF
• On the Exploitability of Instruction Tuning (Jun 2023)  Link  PDF
• LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention (Jun 2023)  Link  PDF

Parameter Efficient Fine Tuning

• Few-Shot Parameter-Efficient Fine-Tuning is Better and Cheaper than In-Context Learning (Aug 2022)  Link  PDF
• The Power of Scale for Parameter-Efficient Prompt Tuning (Sep 2021)  Link  PDF
• LoRA: Low-Rank Adaptation of Large Language Models (Oct 2021)  Link  PDF
• QLoRA: Efficient Finetuning of Quantized LLMs (May 2023)  Link  PDF
• Parameter-Efficient Fine-Tuning without Introducing New Latency (May 2023)  Link  PDF

Reinforcement Learning

• Training language models to follow instructions with human feedback (Mar 2022)  Link  PDF
• Secrets of RLHF in Large Language Models Part I: PPO (Jul 2023)  Link  PDF
• RLAIF: Scaling Reinforcement Learning from Human Feedback with AI Feedback (Sep 2023)  Link  PDF
• Constitutional AI: Harmlessness from AI Feedback (Dec 2022)  Link  PDF
• Fine-tuning Language Models with Generative Adversarial Feedback (Oct 2023)  Link  PDF