{"url":"/task/rag","name":"RAG","slug":"rag","description_markdown":"Retrieval-Augmented Generation (RAG) is a task that combines the strengths of both retrieval-based models and generation-based models. In this approach, a retrieval system selects relevant documents or passages from a large corpus, and a generation model, typically a neural language model, uses the retrieved information to generate a response. This method enhances the accuracy and coherence of generated text, especially in tasks requiring detailed knowledge or long context handling.\r\n\r\nRAG is particularly useful in open-domain question answering, knowledge-grounded dialogue, and summarization tasks. The retrieval step helps the model to access and incorporate external information, making it less reliant on memorized knowledge and better suited for generating responses based on the latest or domain-specific information.\r\n\r\nThe performance of RAG systems is usually measured using metrics such as precision, recall, F1 score, BLEU score, and exact match. Some popular datasets for evaluating RAG models include Natural Questions, MS MARCO, TriviaQA, and SQuAD.","categories":[{"name":"Knowledge Base","url":"/area/knowledge-base"},{"name":"Methodology","url":"/area/methodology"},{"name":"Natural Language Processing","url":"/area/natural-language-processing"}],"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","slug_source":"archive_url"},"counts":{"papers_tagged":2111,"papers_with_code":758,"benchmarks":1,"benchmark_tables_in_archive":1,"benchmark_tables_shown":1,"benchmark_tables_withheld_as_spam":0,"benchmark_definition":"a leaderboard table with at least one row; benchmark_tables_shown also counts the zero-row tables; benchmark_tables_in_archive adds the tables withheld as spam","datasets":6,"subtasks":0,"parent_tasks":1},"benchmarks":[{"leaderboard":"/sota/rag-on-pubmedqa-corpus-with-metadata","slug":"rag-on-pubmedqa-corpus-with-metadata","dataset":"PubMedQA corpus with metadata","dataset_url":"/dataset/pubmedqa-corpus-with-metadata","rows_in_archive":1,"metrics":["ANS-EM"],"first_row_in_archive_order":{"model":"MetaGen Blended RAG","paper_title":"MetaGen Blended RAG: Higher Accuracy for Domain-Specific Q&A Without Fine-Tuning","paper_url":"/paper/metagen-blended-rag-higher-accuracy-for","paper_date":"2025-05-23","arxiv_id":"2505.18247","code_links":[{"title":"ibm-self-serve-assets/metagen-blended-rag","url":"https://github.com/ibm-self-serve-assets/metagen-blended-rag"}],"syntology":null}}],"datasets":[{"url":"/dataset/peerqa","name":"PeerQA","full_name":"","num_papers_in_archive":12},{"url":"/dataset/compmix-ir","name":"CompMix-IR","full_name":"","num_papers_in_archive":1},{"url":"/dataset/crsb","name":"CRSB","full_name":"Context Retrieval Supervision Benchmark","num_papers_in_archive":1},{"url":"/dataset/frames-part","name":"Frames (part)","full_name":"","num_papers_in_archive":1},{"url":"/dataset/pubmedqa-corpus-with-metadata","name":"PubMedQA corpus with metadata","full_name":"","num_papers_in_archive":1},{"url":"/dataset/riskdata","name":"RiskData","full_name":"","num_papers_in_archive":1}],"subtasks":[],"parent_tasks":[{"url":"/task/large-language-model","name":"Large Language Model"}],"papers":{"order":"repositories listed in the archive (desc), then date (desc); the archive holds no stars","population":"papers tagged with this task that list at least one repository in the archive","shown":30,"of":758,"tagged_in_all":2111,"items":[{"url":"/paper/retrieval-augmented-generation-for-knowledge","title":"Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks","date":"2020-05-22","arxiv_id":"2005.11401","repositories_listed":18,"syntology":{"n":6,"n_ran":4,"n_unverified":2,"n_pointer_only":0}},{"url":"/paper/self-rag-learning-to-retrieve-generate-and","title":"Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection","date":"2023-10-17","arxiv_id":"2310.11511","repositories_listed":6,"syntology":{"n":14,"n_ran":8,"n_unverified":6,"n_pointer_only":3}},{"url":"/paper/r1-searcher-incentivizing-the-search","title":"R1-Searcher: Incentivizing the Search Capability in LLMs via Reinforcement Learning","date":"2025-03-07","arxiv_id":"2503.05592","repositories_listed":5,"syntology":{"n":1,"n_ran":1,"n_unverified":0,"n_pointer_only":0}},{"url":"/paper/colpali-efficient-document-retrieval-with","title":"ColPali: Efficient Document Retrieval with Vision Language Models","date":"2024-06-27","arxiv_id":"2407.01449","repositories_listed":5,"syntology":{"n":4,"n_ran":2,"n_unverified":2,"n_pointer_only":0}},{"url":"/paper/retrieval-augmented-generation-for-large","title":"Retrieval-Augmented Generation for Large Language Models: A Survey","date":"2023-12-18","arxiv_id":"2312.10997","repositories_listed":4,"syntology":null},{"url":"/paper/simpledeepsearcher-deep-information-seeking","title":"SimpleDeepSearcher: Deep Information Seeking via Web-Powered Reasoning Trajectory Synthesis","date":"2025-05-22","arxiv_id":"2505.16834","repositories_listed":3,"syntology":{"n":2,"n_ran":2,"n_unverified":0,"n_pointer_only":0}},{"url":"/paper/r1-searcher-incentivizing-the-dynamic","title":"R1-Searcher++: Incentivizing the Dynamic Knowledge Acquisition of LLMs via Reinforcement Learning","date":"2025-05-22","arxiv_id":"2505.17005","repositories_listed":3,"syntology":{"n":10,"n_ran":1,"n_unverified":9,"n_pointer_only":0}},{"url":"/paper/search-r1-training-llms-to-reason-and","title":"Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning","date":"2025-03-12","arxiv_id":"2503.09516","repositories_listed":3,"syntology":{"n":2,"n_ran":2,"n_unverified":0,"n_pointer_only":1}},{"url":"/paper/lrp4rag-detecting-hallucinations-in-retrieval","title":"LRP4RAG: Detecting Hallucinations in Retrieval-Augmented Generation via Layer-wise Relevance Propagation","date":"2024-08-28","arxiv_id":"2408.15533","repositories_listed":3,"syntology":null},{"url":"/paper/from-local-to-global-a-graph-rag-approach-to","title":"From Local to Global: A Graph RAG Approach to Query-Focused Summarization","date":"2024-04-24","arxiv_id":"2404.16130","repositories_listed":3,"syntology":{"n":5,"n_ran":0,"n_unverified":5,"n_pointer_only":0}},{"url":"/paper/retrieval-augmented-generation-for-ai","title":"Retrieval-Augmented Generation for AI-Generated Content: A Survey","date":"2024-02-29","arxiv_id":"2402.19473","repositories_listed":3,"syntology":null},{"url":"/paper/the-power-of-noise-redefining-retrieval-for","title":"The Power of Noise: Redefining Retrieval for RAG Systems","date":"2024-01-26","arxiv_id":"2401.14887","repositories_listed":3,"syntology":{"n":5,"n_ran":5,"n_unverified":0,"n_pointer_only":5}},{"url":"/paper/ragtruth-a-hallucination-corpus-for","title":"RAGTruth: A Hallucination Corpus for Developing Trustworthy Retrieval-Augmented Language Models","date":"2023-12-31","arxiv_id":"2401.00396","repositories_listed":3,"syntology":{"n":5,"n_ran":3,"n_unverified":2,"n_pointer_only":0}},{"url":"/paper/ragas-automated-evaluation-of-retrieval","title":"RAGAS: Automated Evaluation of Retrieval Augmented Generation","date":"2023-09-26","arxiv_id":"2309.15217","repositories_listed":3,"syntology":{"n":6,"n_ran":0,"n_unverified":6,"n_pointer_only":0}},{"url":"/paper/robust-affine-feature-matching-via-quadratic","title":"Robust affine point matching via quadratic assignment on Grassmannians","date":"2023-03-05","arxiv_id":"2303.02698","repositories_listed":3,"syntology":null},{"url":"/paper/arctic-long-sequence-training-scalable-and","title":"Arctic Long Sequence Training: Scalable And Efficient Training For Multi-Million Token Sequences","date":"2025-06-16","arxiv_id":"2506.13996","repositories_listed":2,"syntology":null},{"url":"/paper/a-survey-of-llm-times-data","title":"A Survey of LLM $\\times$ DATA","date":"2025-05-24","arxiv_id":"2505.18458","repositories_listed":2,"syntology":null},{"url":"/paper/relevance-isn-t-all-you-need-scaling-rag","title":"Relevance Isn't All You Need: Scaling RAG Systems With Inference-Time Compute Via Multi-Criteria Reranking","date":"2025-03-14","arxiv_id":"2504.07104","repositories_listed":2,"syntology":null},{"url":"/paper/benchmarking-retrieval-augmented-generation-1","title":"Benchmarking Retrieval-Augmented Generation in Multi-Modal Contexts","date":"2025-02-24","arxiv_id":"2502.17297","repositories_listed":2,"syntology":null},{"url":"/paper/lettucedetect-a-hallucination-detection","title":"LettuceDetect: A Hallucination Detection Framework for RAG Applications","date":"2025-02-24","arxiv_id":"2502.17125","repositories_listed":2,"syntology":null},{"url":"/paper/infinite-retrieval-attention-enhanced-llms-in","title":"Infinite Retrieval: Attention Enhanced LLMs in Long-Context Processing","date":"2025-02-18","arxiv_id":"2502.12962","repositories_listed":2,"syntology":null},{"url":"/paper/ras-retrieval-and-structuring-for-knowledge","title":"RAS: Retrieval-And-Structuring for Knowledge-Intensive LLM Generation","date":"2025-02-16","arxiv_id":"2502.10996","repositories_listed":2,"syntology":{"n":13,"n_ran":1,"n_unverified":12,"n_pointer_only":2}},{"url":"/paper/autoagent-a-fully-automated-and-zero-code","title":"AutoAgent: A Fully-Automated and Zero-Code Framework for LLM Agents","date":"2025-02-09","arxiv_id":"2502.05957","repositories_listed":2,"syntology":null},{"url":"/paper/webwalker-benchmarking-llms-in-web-traversal","title":"WebWalker: Benchmarking LLMs in Web Traversal","date":"2025-01-13","arxiv_id":"2501.07572","repositories_listed":2,"syntology":null},{"url":"/paper/eliza-a-web3-friendly-ai-agent-operating","title":"Eliza: A Web3 friendly AI Agent Operating System","date":"2025-01-12","arxiv_id":"2501.06781","repositories_listed":2,"syntology":null},{"url":"/paper/search-o1-agentic-search-enhanced-large","title":"Search-o1: Agentic Search-Enhanced Large Reasoning Models","date":"2025-01-09","arxiv_id":"2501.05366","repositories_listed":2,"syntology":null},{"url":"/paper/jasper-and-stella-distillation-of-sota","title":"Jasper and Stella: distillation of SOTA embedding models","date":"2024-12-26","arxiv_id":"2412.19048","repositories_listed":2,"syntology":{"n":4,"n_ran":1,"n_unverified":3,"n_pointer_only":0}},{"url":"/paper/initial-nugget-evaluation-results-for-the","title":"Initial Nugget Evaluation Results for the TREC 2024 RAG Track with the AutoNuggetizer Framework","date":"2024-11-14","arxiv_id":"2411.09607","repositories_listed":2,"syntology":null},{"url":"/paper/autorag-automated-framework-for-optimization","title":"AutoRAG: Automated Framework for optimization of Retrieval Augmented Generation Pipeline","date":"2024-10-28","arxiv_id":"2410.20878","repositories_listed":2,"syntology":null},{"url":"/paper/lorann-low-rank-matrix-factorization-for","title":"LoRANN: Low-Rank Matrix Factorization for Approximate Nearest Neighbor Search","date":"2024-10-24","arxiv_id":"2410.18926","repositories_listed":2,"syntology":{"n":20,"n_ran":0,"n_unverified":20,"n_pointer_only":0}}],"syntology_records":14,"syntology_note":"a paper without a record is not a recorded non-run: it may lack an arXiv id or simply be absent from the graph layer"},"description_links":{"kept":0,"unwrapped_to_text":0,"bare_urls_linked":0,"relative_images_dropped":0,"rule":"internal links are kept only when the target slug exists in the catalog"},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per-sample execution status on synthesized fixtures ('ran N of M samples'); not a correctness claim and not a ranking signal.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"}},"not_shown":{"libraries":"the archive has no per-task library table","trend_sparklines":"the Trend column of the benchmarks table was a rendered image; it is not in the archive","social_and_latest_sorts":"stars and social signals are not in the archive"}}