# RAG Resources - [Files from my "Using Retrieval Augmented Generation (RAG), Langchain, and LLMs for Cybersecurity Operations" Course](https://github.com/santosomar/RAG-for-cybersecurity) - [LangFlow](https://github.com/langflow-ai/langflow) - a low-code app builder for RAG and multi-agent AI applications. It’s Python-based and agnostic to any model, API, or database. ### Disadvantages of RAG - [Disadvantages of RAG](https://medium.com/@kelvin.lu.au/disadvantages-of-rag-5024692f2c53) ### RAG Patterns - [Generative AI Lifecycle Patterns](https://dr-arsanjani.medium.com/the-generative-ai-lifecycle-1b0c7d9463ec) - [Why do RAG pipelines fail? Advanced RAG Patterns — Part1 Ozgur Guler](https://cloudatlas.me/why-do-rag-pipelines-fail-advanced-rag-patterns-part1-841faad8b3c2) - [How to improve RAG peformance — Advanced RAG Patterns — Part2](https://cloudatlas.me/how-to-improve-rag-peformance-advanced-rag-patterns-part2-0c84e2df66e6) - [Patterns for Building LLM-based Systems & Products](https://eugeneyan.com/writing/llm-patterns/) - [AI Engineer Summit - Building Blocks for LLM Systems & Products](https://eugeneyan.com/speaking/ai-eng-summit/) - [Technical Considerations for Complex RAG](https://medium.com/enterprise-rag/a-first-intro-to-complex-rag-retrieval-augmented-generation-a8624d70090f) ### Dialogue Routing - [Routing in RAG-Driven Applications](https://towardsdatascience.com/routing-in-rag-driven-applications-a685460a7220) ## Retrieval ### Vector Retrieval - [Boosting RAG: Picking the Best Embedding & Reranker models](https://blog.llamaindex.ai/boosting-rag-picking-the-best-embedding-reranker-models-42d079022e83) - [What We Need to Know Before Adopting a Vector Database](https://medium.com/@kelvin.lu.au/what-we-need-to-know-before-adopting-a-vector-database-85e137570fbb) #### Chunking - [Chunking Strategies for LLM Applications](https://www.pinecone.io/learn/chunking-strategies/) - [Evaluating the Ideal Chunk Size for a RAG System using LlamaIndex](https://blog.llamaindex.ai/evaluating-the-ideal-chunk-size-for-a-rag-system-using-llamaindex-6207e5d3fec5) - [How to Chunk Text Data — A Comparative Analysis](https://towardsdatascience.com/how-to-chunk-text-data-a-comparative-analysis-3858c4a0997a) #### Embeddings #### Vector Search - [Awesome Search](https://github.com/frutik/awesome-search) - [Advanced RAG Retrieval Strategies: Sentence Window Retrieval](https://generativeai.pub/advanced-rag-retrieval-strategies-sentence-window-retrieval-b6964b6e56f7) ### Not Vector Retrieval - [Vector Search Is Not All You Need](https://towardsdatascience.com/vector-search-is-not-all-you-need-ecd0f16ad65e) - [Build a search engine, not a vector DB](https://blog.elicit.com/search-vs-vector-db/) - [Improving RAG (Retrieval Augmented Generation) Answer Quality with Re-ranker](https://medium.com/towards-generative-ai/improving-rag-retrieval-augmented-generation-answer-quality-with-re-ranker-55a19931325) - [From Search to Synthesis: Enhancing RAG with BM25 and Reciprocal Rank Fusion](https://medium.com/@kachari.bikram42/from-search-to-synthesis-enhancing-rag-with-bm25-and-reciprocal-rank-fusion-872d21dc4ca7) ## Generation ### Prompts - [Emerging RAG & Prompt Engineering Architectures for LLMs](https://cobusgreyling.medium.com/updated-emerging-rag-prompt-engineering-architectures-for-llms-17ee62e5cbd9) - [How to Cut RAG Costs by 80% Using Prompt Compression](https://towardsdatascience.com/how-to-cut-rag-costs-by-80-using-prompt-compression-877a07c6bedb) #### Prompting strategies ##### Multi-Modal RAG - [Multi-Modal RAG](https://blog.llamaindex.ai/multi-modal-rag-621de7525fea) ##### Multi-index RAG - [Having all of your data stored in one collection isn't always the best for RAG apps](https://twitter.com/ecardenas300/status/1724829560041038072) ##### Multi-Document - [Advanced RAG — Multi-Documents Agent with LlamaIndex](https://blog.gopenai.com/advanced-rag-multi-documents-agent-with-llamaindex-43b604f84909) ##### FLARE - [Better RAG with Active Retrieval Augmented Generation FLARE](https://blog.lancedb.com/better-rag-with-active-retrieval-augmented-generation-flare-3b66646e2a9f) ##### Chain-of-Verification - [in-Of-Verification Reduces Hallucination in LLMs](https://cobusgreyling.medium.com/chain-of-verification-reduces-hallucination-in-llms-20af5ea67672) ##### Chain-Of-Thought - [Chain-Of-Thought Prompting In LLMs](https://cobusgreyling.medium.com/chain-of-thought-prompting-in-llms-1077164edf97) ### Context - [The Needle In a Haystack Test](https://towardsdatascience.com/the-needle-in-a-haystack-test-a94974c1ad38) - [Conversational Memory for LLMs with Langchain](https://www.pinecone.io/learn/series/langchain/langchain-conversational-memory/) #### Long context RAG - [The next generation of RAG: Long-Context RAG](https://twitter.com/ecardenas300/status/1724129722492142048) - [NVIDIA Research: RAG with Long Context LLMs](https://blog.llamaindex.ai/nvidia-research-rag-with-long-context-llms-7d94d40090c4) #### Knowledge and Knowledge Graphs - [Graph RAG: Unleashing the Power of Knowledge Graphs with LLM](https://medium.com/@nebulagraph/graph-rag-the-new-llm-stack-with-knowledge-graphs-e1e902c504ed) - [Embeddings + Knowledge Graphs: The Ultimate Tools for RAG Systems](https://towardsdatascience.com/embeddings-knowledge-graphs-the-ultimate-tools-for-rag-systems-cbbcca29f0fd) - [The Practical Benefits to Grounding an LLM in a Knowledge Graph Daniel Bukowski](https://medium.com/@bukowski.daniel/the-practical-benefits-to-grounding-an-llm-in-a-knowledge-graph-919918eb493) - [Implement RAG with Knowledge Graph and Llama-Index](https://medium.aiplanet.com/implement-rag-with-knowledge-graph-and-llama-index-6a3370e93cdd) - [Awesome Knowledge Graphs](https://github.com/frutik/awesome-knowledge-graphs) - [HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models](https://arxiv.org/abs/2405.14831) ### Automated prompt optimization ### Hallucination - [How to Detect Hallucinations in LLMs](https://towardsdatascience.com/real-time-llm-hallucination-detection-9a68bb292698) - [Measuring Hallucinations in RAG Systems](https://vectara.com/measuring-hallucinations-in-rag-systems/) ### Guardrails - [Safeguarding LLMs with Guardrails](https://towardsdatascience.com/safeguarding-llms-with-guardrails-4f5d9f57cff2) - [NeMo Guardrails: The Missing Manual](https://www.pinecone.io/learn/nemo-guardrails-intro/) ## LLM Models ### Finetuning and Pretraining - [Fine-Tuning Llama 2.0 with Single GPU Magic](https://ai.plainenglish.io/fine-tuning-llama2-0-with-qloras-single-gpu-magic-1b6a6679d436) - [Practitioners guide to fine-tune LLMs for domain-specific use case](https://cismography.medium.com/practitioners-guide-to-fine-tune-llms-for-domain-specific-use-case-part-1-4561714d874f) - [Are You Pre-training your RAG Models on Your Raw Text?](https://medium.com/thirdai-blog/are-you-pre-training-your-rag-models-on-your-raw-text-40f832d87703) - [Combine Multiple LoRA Adapters for Llama 2](https://towardsdatascience.com/combine-multiple-lora-adapters-for-llama-2-ea0bef9025cf) - [RAG vs Finetuning — Which Is the Best Tool to Boost Your LLM Application?](https://towardsdatascience.com/rag-vs-finetuning-which-is-the-best-tool-to-boost-your-llm-application-94654b1eaba7) ## Evaluation of RAGs - [RAG Evaluation](https://cobusgreyling.medium.com/rag-evaluation-9813a931b3d4) - [Evaluating RAG: A journey through metrics](https://www.elastic.co/search-labs/blog/articles/evaluating-rag-metrics) - [Exploring End-to-End Evaluation of RAG Pipelines](https://betterprogramming.pub/exploring-end-to-end-evaluation-of-rag-pipelines-e4c03221429) - [Evaluation Driven Development, the Swiss Army Knife for RAG Pipelines](https://levelup.gitconnected.com/evaluation-driven-development-the-swiss-army-knife-for-rag-pipelines-dba24218d47e) - [Evaluating the Ideal Chunk Size for a RAG System using LlamaIndex](https://blog.llamaindex.ai/evaluating-the-ideal-chunk-size-for-a-rag-system-using-llamaindex-6207e5d3fec5) ## Performance and cost - [Secrets to Optimizing RAG LLM Apps for Better Performance, Accuracy and Lower Costs!](https://medium.com/madhukarkumar/secrets-to-optimizing-rag-llm-apps-for-better-accuracy-performance-and-lower-cost-da1014127c0a) ## Privacy - [Masking PII Data in RAG Pipeline](https://betterprogramming.pub/masking-pii-data-in-rag-pipeline-326d2d330336) ## Security - [Hijacking Chatbots: Dangerous Methods Manipulating GPTs](https://medium.com/@jankammerath/hijacking-chatbots-dangerous-methods-manipulating-gpts-52342f4f88b8) ## Applications of RAG ### Chatbots ## Tools - [Three Open-Source RAG Tools You Need to Know About](https://medium.com/programmers-journey/three-open-source-rag-tools-you-need-to-know-about-331c3f28ab22) - [HayStack](https://github.com/deepset-ai/haystack) - [RAGAS](https://github.com/explodinggradients/ragas) ### DSPy - [DSPy — Does It Live Up To The Hype?](https://medium.com/emalpha/dspy-does-it-live-up-to-the-hype-6e56c2c6e7a0) ### AutoRAG - [AutoRAG](https://github.com/Marker-Inc-Korea/AutoRAG) - AutoML tool for RAG. Automatically optimize RAG pipeline with single YAML file. ### AutoGPT ### Langchain - [Langchain](https://github.com/langchain-ai/langchain) - [Langchain is NOT for production use. Here is why ..](https://medium.com/@aldendorosario/langchain-is-not-for-production-use-here-is-why-9f1eca6cce80) ### LlamaIndex - [LlamaIndex](https://github.com/run-llama/llama_index) - [Building Production-Ready LLM Apps with LlamaIndex: Document Metadata for Higher Accuracy Retrieval](https://betterprogramming.pub/building-production-ready-llm-apps-with-llamaindex-document-metadata-for-higher-accuracy-retrieval-a8ceca641fb5) - [Building Production-Ready LLM Apps With LlamaIndex: Recursive Document Agents for Dynamic Retrieval](https://betterprogramming.pub/building-production-ready-llm-apps-with-llamaindex-recursive-document-agents-for-dynamic-retrieval-1f4b25287918) ## Vendor-specific examples - [RAG Pipeline with Mistral 7B Instruct Model in Colab: A Step-by-Step Guide Qendel AI GoPenAI](https://blog.gopenai.com/rag-pipeline-with-mistral-7b-instruct-model-a-step-by-step-guide-138df378a0c2) ### Elastcisearch + OpenAI - [ChatGPT and Elasticsearch: OpenAI meets private data](https://www.elastic.co/search-labs/blog/chatgpt-elasticsearch-openai-meets-private-data) ### OpenAI and ChatGPT - [Compare PDF Question Answering Systems Build with OpenAI and Google VertexAI](https://medium.com/@kelvin.lu.au/compare-pdf-question-answering-with-openai-and-google-vertexai-46638d62327b) #### Tools and fucntions - [Unlocking the Power of the OpenAI API: Master Function-Calling with Practical Examples](https://medium.com/@apollovro/unlocking-the-power-of-the-openai-api-master-function-calling-with-practical-examples-f8b9ab2fceec) - [penAI/Chat-GPT Function Calling : for Enhanced AI Interactions](https://levelup.gitconnected.com/openai-chat-gpt-function-calling-for-enhanced-ai-interactions-338be974027) ### Vespa - [Hands-On RAG guide for personal data with Vespa and LLamaIndex](https://blog.vespa.ai/scaling-personal-ai-assistants-with-streaming-mode/) ### Qdrant ## Running RAGs in production ## Vectors corner - [Similarity Search, Part 2: Product Quantization](https://towardsdatascience.com/similarity-search-product-quantization-b2a1a6397701) - [Binary and Scalar Embedding Quantization for Significantly Faster & Cheaper Retrieval](https://huggingface.co/blog/embedding-quantization) - [Cohere int8 & binary Embeddings - Scale Your Vector Database to Large Datasets Image of Nils Reimers](https://cohere.com/blog/int8-binary-embeddings) ## Research Papers ### Survey and Benchmark **Benchmarking Large Language Models in Retrieval-Augmented Generation** \ *Jiawei Chen, Hongyu Lin, Xianpei Han, Le Sun* \ arXiv 2023. [[Paper](https://arxiv.org/abs/2309.01431)][[Github](https://github.com/chen700564/RGB)] \ 4 Sep 2023 ### Retrieval-enhanced LLMs **Chain-of-Note: Enhancing Robustness in Retrieval-Augmented Language Models** \ *Wenhao Yu, Hongming Zhang, Xiaoman Pan, Kaixin Ma, Hongwei Wang, Dong Yu* \ arxiv - Nov 2023 [[Paper](https://arxiv.org/abs/2311.09210)] **REST: Retrieval-Based Speculative Decoding** \ *Zhenyu He, Zexuan Zhong, Tianle Cai, Jason D Lee, Di He* \ arXiv - Nov 2023 [[Paper](https://arxiv.org/abs/2311.08252)][[Github](https://github.com/fasterdecoding/rest)] **Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection** *Anonymous* ICLR 24 – Oct 2023 [[paper](https://openreview.net/forum?id=hSyW5go0v8)] **Self-Knowledge Guided Retrieval Augmentation for Large Language Models** \ *Yile Wang, Peng Li, Maosong Sun, Yang Liu* \ arXiv - Oct 2023 [[Ppaer](https://arxiv.org/abs/2310.05002)] **Retrieval meets Long Context Large Language Models** \ *Peng Xu, Wei Ping, Xianchao Wu, Lawrence McAfee, Chen Zhu, Zihan Liu, Sandeep Subramanian, Evelina Bakhturina, Mohammad Shoeybi, Bryan Catanzaro* \ arxiv - Oct 2023 [[Paper](https://arxiv.org/abs/2310.03025)] **DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines** *Omar Khattab, Arnav Singhvi, Paridhi Maheshwari, Zhiyuan Zhang, Keshav Santhanam, Sri Vardhamanan, Saiful Haq, Ashutosh Sharma, Thomas T. Joshi, Hanna Moazam, Heather Miller, Matei Zaharia, Christopher Potts* arXiv – Oct 2023 [[paper](https://arxiv.org/abs/2310.03714)] [[code](https://github.com/stanfordnlp/dspy)] **Adaptive Chameleon or Stubborn Sloth: Revealing the Behavior of Large Language Models in Knowledge Conflicts** *Jian Xie, Kai Zhang, Jiangjie Chen, Renze Lou, Yu Su* ICLR 24 – May 2023 [[paper](https://arxiv.org/abs/2305.13300)] [[code](https://github.com/OSU-NLP-Group/LLM-Knowledge-Conflict)] **Active Retrieval Augmented Generation** *Zhengbao Jiang, Frank F. Xu, Luyu Gao, Zhiqing Sun, Qian Liu, Jane Dwivedi-Yu, Yiming Yang, Jamie Callan, Graham Neubig* arXiv – May 2023 [[paper](https://arxiv.org/abs/2305.06983)] [[code](https://github.com/jzbjyb/FLARE)] **REPLUG: Retrieval-Augmented Black-Box Language Models** *Weijia Shi, Sewon Min, Michihiro Yasunaga, Minjoon Seo, Rich James, Mike Lewis, Luke Zettlemoyer, Wen-tau Yih* arXiv – Jan 2023 [[paper](https://arxiv.org/abs/2301.12652)] **Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks** *Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, Sebastian Riedel, Douwe Kiela* NeurIPS 2020 - May 2020 [[Paper](https://arxiv.org/abs/2005.11401)] ### RAG Instruction Tuning **RA-DIT: Retrieval-Augmented Dual Instruction Tuning** *Anonymous* ICLR 24 – Oct 23 [[paper](https://openreview.net/forum?id=22OTbutug9)] **InstructRetro: Instruction Tuning post Retrieval-Augmented Pretraining** *Boxin Wang, Wei Ping, Lawrence McAfee, Peng Xu, Bo Li, Mohammad Shoeybi, Bryan Catanzaro* \ arXiv - Oct 23 [[paper](https://openreview.net/forum?id=4stB7DFLp6)] ### RAG In-Context Learning **In-Context Retrieval-Augmented Language Models** *Ori Ram, Yoav Levine, Itay Dalmedigos, Dor Muhlgay, Amnon Shashua, Kevin Leyton-Brown, Yoav Shoham* AI21 Labs – Jan 2023 [[paper](https://uploads-ssl.webflow.com/60fd4503684b466578c0d307/63c6c20dec4479564db21819_NEW_In_Context_Retrieval_Augmented_Language_Models.pdf)] [[code](https://github.com/AI21Labs/in-context-ralm)] ### RAG Embeddings **RegaVAE: A Retrieval-Augmented Gaussian Mixture Variational Auto-Encoder for Language Modeling** \ *Jingcheng Deng, Liang Pang, Huawei Shen, Xueqi Cheng* \ EMNLP 2023 - Oct 2023 [[Paper](https://arxiv.org/abs/2310.10567)][[Github](https://github.com/TrustedLLM/RegaVAE)] **Text Embeddings Reveal (Almost) As Much As Text** \ *John X. Morris, Volodymyr Kuleshov, Vitaly Shmatikov, Alexander M. Rush* \ EMNLP 2023 - Oct 2023 [[Paper](https://arxiv.org/abs/2310.06816?ref=upstract.com)][[Github](https://github.com/jxmorris12/vec2text)] **Jina Embeddings 2: 8192-Token General-Purpose Text Embeddings for Long Documents** \ *Michael Günther, Jackmin Ong, Isabelle Mohr, Alaeddine Abdessalem, Tanguy Abel, Mohammad Kalim Akram, Susana Guzman, Georgios Mastrapas, Saba Sturua, Bo Wang, Maximilian Werk, Nan Wang, Han Xiao* \ arXiv - Oct 2023. [[Paper](https://arxiv.org/abs/2310.19923)][[Model](https://huggingface.co/jinaai/jina-embeddings-v2-small-en)] ### RAG Simulators **KAUCUS: Knowledge Augmented User Simulators for Training Language Model Assistants** \ *Kaustubh D. Dhole* \ Simulation of Conversational Intelligence in Chat, EACL 2024 [[Paper](https://arxiv.org/abs/2401.16454)] ### RAG Search ### RAG Long-text and Memory **HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models** \ *Bernal Jiménez Gutiérrez, Yiheng Shu, Yu Gu, Michihiro Yasunaga, Yu Su* \ arXiv - May 2024 [[paper](https://arxiv.org/abs/2405.14831)] [[GitHub](https://github.com/OSU-NLP-Group/HippoRAG)] **Understanding Retrieval Augmentation for Long-Form Question Answering** \ *Hung-Ting Chen, Fangyuan Xu, Shane A. Arora, Eunsol Choi* \ arXiv - Oct 2023 [[Paper](https://arxiv.org/abs/2310.12150)] ### RAG Evaluation **ARES: An Automated Evaluation Framework for Retrieval-Augmented Generation Systems** \ *Jon Saad-Falcon, Omar Khattab, Christopher Potts, Matei Zaharia* \ arXiv - Nov 2023. [[Paper](https://arxiv.org/abs/2311.09476)] [[Github](https://github.com/stanford-futuredata/ares)] ### RAG Optimization **Learning to Filter Context for Retrieval-Augmented Generation** \ *Zhiruo Wang, Jun Araki, Zhengbao Jiang, Md Rizwan Parvez, Graham Neubig* \ arxiv- Nov 2023 [[Paper](https://arxiv.org/abs/2311.08377)][[Github](https://github.com/zorazrw/filco)] **Large Language Models Can Be Easily Distracted by Irrelevant Context** \ *Freda Shi, Xinyun Chen, Kanishka Misra, Nathan Scales, David Dohan, Ed Chi, Nathanael Schärli, Denny Zhou* \ ICML 2023 - Jan 2023 [[Paper](https://arxiv.org/abs/2302.00093)][[Github](https://github.com/google-research-datasets/GSM-IC)] **Evidentiality-guided Generation for Knowledge-Intensive NLP Tasks** \ *Akari Asai, Matt Gardner, Hannaneh Hajishirzi* \ NAACL 2022 - Dec 2021 [[Paper](https://arxiv.org/abs/2112.08688)][[Github](https://github.com/akariasai/evidentiality_qa)] **When Not to Trust Language Models: Investigating Effectiveness of Parametric and Non-Parametric Memories** \ *Alex Mallen, Akari Asai, Victor Zhong, Rajarshi Das, Daniel Khashabi, Hannaneh Hajishirzi* \ ACL 2023 - Dec 2022 [[Paper](https://arxiv.org/abs/2212.10511)][[Github](https://github.com/alextmallen/adaptive-retrieval)] ### RAG Application **Deficiency of Large Language Models in Finance: An Empirical Examination of Hallucination** \ *Haoqiang Kang, Xiao-Yang Liu* \ arXiv - Nov 2023 [[Paper](https://arxiv.org/abs/2311.15548)] **Clinfo.ai: An Open-Source Retrieval-Augmented Large Language Model System for Answering Medical Questions using Scientific Literature** \ *Alejandro Lozano, Scott L Fleming, Chia-Chun Chiang, Nigam Shah* \ arXiv - Oct 2023. [[Paper](https://arxiv.org/abs/2310.16146v1)] **PEARL: Personalizing Large Language Model Writing Assistants with Generation-Calibrated Retrievers** \ *Sheshera Mysore, Zhuoran Lu, Mengting Wan, Longqi Yang, Steve Menezes, Tina Baghaee, Emmanuel Barajas Gonzalez, Jennifer Neville, Tara Safavi* \ arXiv - Nov 2023. [[Paper](https://arxiv.org/abs/2311.09180)]