Skip to content

Latest commit

 

History

History
241 lines (164 loc) · 19.3 KB

README.md

File metadata and controls

241 lines (164 loc) · 19.3 KB

Awesome RAG

A curated list of retrieval-augmented generation (RAG) in large language models.
Suggest and discuss possible enhancements on the Potential Additions page.

cc

Table of Content:

Papers

Survey

2024

  • Retrieval-Augmented Generation for Large Language Models: A Survey
    Paper Code arXiv Dynamic JSON Badge

  • Retrieval-Augmented Generation for Natural Language Processing: A Survey
    Paper arXiv Dynamic JSON Badge

2023

  • Benchmarking Large Language Models in Retrieval-Augmented Generation
    Paper arXiv Dynamic JSON Badge

2022

  • A Survey on Retrieval-Augmented Text Generation
    Paper arXiv Dynamic JSON Badge

General

2024

  • Learning to Retrieve In-Context Examples for Large Language Models
    Paper Code EACL Dynamic JSON Badge

  • Reliable, Adaptable, and Attributable Language Models with Retrieval
    Paper arXiv Dynamic JSON Badge

2023

  • Active Retrieval Augmented Generation
    Paper Code EMNLP Architecture Dynamic JSON Badge

  • REPLUG: Retrieval-Augmented Black-Box Language Models
    Paper arXiv Architecture Dynamic JSON Badge

  • Shall We Pretrain Autoregressive Language Models with Retrieval? A Comprehensive Study
    Paper Code EMNLP Dynamic JSON Badge

  • InstructRetro: Instruction Tuning post Retrieval-Augmented Pretraining
    Paper Code arXiv Dynamic JSON Badge

  • Retrieve Anything To Augment Large Language Models
    Paper Code arXiv Dynamic JSON Badge

  • Reimagining Retrieval Augmented Language Models for Answering Queries
    Paper ACL Dynamic JSON Badge

  • In-Context Retrieval-Augmented Language Models
    Paper Code TACL Architecture Dynamic JSON Badge

  • Query Rewriting for Retrieval-Augmented Large Language Models
    Paper Code EMNLP Dynamic JSON Badge

  • Pre-computed memory or on-the-fly encoding? A hybrid approach to retrieval augmentation makes the most of your compute
    Paper PMLR Dynamic JSON Badge

  • Universal Information Extraction with Meta-Pretrained Self-Retrieval
    Paper Code ACL Dynamic JSON Badge

  • RAVEN: In-Context Learning with Retrieval Augmented Encoder-Decoder Language Models
    Paper arXiv Dynamic JSON Badge

  • Unlimiformer: Long-Range Transformers with Unlimited Length Input
    Paper Code NeurIPS Architecture Dynamic JSON Badge

  • Nonparametric Masked Language Modeling
    Paper Code ACL Training Dynamic JSON Badge

2022

  • Improving language models by retrieving from trillions of tokens
    Paper Blog PMLR Architecture Dynamic JSON Badge

  • Recitation-Augmented Language Models
    Paper Code ICLR Dynamic JSON Badge

  • Atlas: Few-shot Learning with Retrieval Augmented Language Models
    Paper Code Blog Training JMLR Dynamic JSON Badge

  • You can't pick your neighbors, or can you? When and how to rely on retrieval in the kNN-LM
    Paper ACL Architecture Dynamic JSON Badge

  • Neuro-Symbolic Language Modeling with Automaton-augmented Retrieval
    Paper Code ICML Architecture Dynamic JSON Badge

  • Memorizing Transformers
    Paper Code ICLR Architecture Dynamic JSON Badge

  • Training Language Models with Memory Augmentation
    Paper EMNLP Training Dynamic JSON Badge

  • Unsupervised Dense Information Retrieval with Contrastive Learning
    Paper Code arXiv Training Dynamic JSON Badge

  • Teaching language models to support answers with verified quotes
    Paper arXiv Application Dynamic JSON Badge

  • kNN-Prompt: Nearest Neighbor Zero-Shot Inference
    Paper Code EMNLP Application Dynamic JSON Badge

2021

  • Efficient Nearest Neighbor Language Models
    Paper Code EMNLP Architecture Dynamic JSON Badge

  • Mention Memory: incorporating textual knowledge into Transformers through entity mention attention
    Paper Code arXiv Architecture Dynamic JSON Badge

2020

  • REALM: Retrieval-Augmented Language Model Pre-Training
    Paper Code HuggingFace PMLR Architecture Dynamic JSON Badge

  • Generalization through Memorization: Nearest Neighbor Language Models
    Paper Code ICLR Architecture Dynamic JSON Badge

  • Entities as Experts: Sparse Memory Access with Entity Supervision
    Paper EMNLP Architecture Dynamic JSON Badge

  • Dense Passage Retrieval for Open-Domain Question Answering
    Paper Code EMNLP Training Dynamic JSON Badge

  • Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks
    Paper HuggingFace NeurIPS Dynamic JSON Badge

Resources

Lectures, Talks, Tutorials

2024

  • Modular RAG and RAG Flow
    Yunfan Gao (2024) Tutorial
    Blog I Blog II

2023

  • Stanford CS25: V3 I Retrieval Augmented Language Models
    Douwe Kiela (2023) Lecture
    Video

  • Building RAG-based LLM Applications for Production
    Anyscale (2023) Tutorial
    Blog

  • Multi-Vector Retriever for RAG on tables, text, and images
    LangChain (2023) Tutorial
    Blog

  • Retrieval-based Language Models and Applications
    Asai et al. (2023) Tutorial ACL
    Website Video

  • Advanced RAG Techniques: an Illustrated Overview
    Ivan Ilin (2023) Tutorial
    Blog

  • Retrieval Augmented Language Modeling
    Melissa Dell (2023) Lecture
    Video

Workshops

2024

  • Towards Knowledgeable Language Models
    Zoey Sha Li, Manling Li, Michael JQ Zhang, Eunsol Choi, Mor Geva, Peter Hase
    @ACL 2024, August 12-17
    Website

Tools

  • LangChain
    LangChain is a framework for developing applications powered by language models.
    Website

  • LlamaIndex
    LlamaIndex is a simple, flexible data framework for connecting custom data sources to large language models.
    Website

  • Verba
    Verba is an open-source application designed to offer an end-to-end, streamlined, and user-friendly interface for Retrieval-Augmented Generation (RAG) out of the box.
    Website

  • NEUM
    Open-source RAG framework optimized for large-scale and real-time data.
    Website

  • Unstructured
    Unstructured.io offers a powerful toolkit that handles the ingestion and data preprocessing step, allowing you to focus on the more exciting downstream steps in your machine learning pipeline. Unstructured has over a dozen data connectors that easily integrate with various data sources, including AWS S3, Discord, Slack, Wikipedia, and more.
    Website

Other Collections