mlsanityData to 7 Oct 2026

Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection

arXiv 2310.11511 ↗First posted 17 Oct 2023cs.CL cs.AI cs.LGcited by 1,800

On the map: Retrieval-augmented generation → RAG robustness and adaptive retrieval

Abstract (arXiv)

Despite their remarkable capabilities, large language models (LLMs) often produce responses containing factual inaccuracies due to their sole reliance on the parametric knowledge they encapsulate. Retrieval-Augmented Generation (RAG), an ad hoc approach that augments LMs with retrieval of relevant knowledge, decreases such issues. However, indiscriminately retrieving and incorporating a fixed number of retrieved passages, regardless of whether retrieval is necessary, or passages are relevant, diminishes LM versatility or can lead to unhelpful response generation. We introduce a new framework called Self-Reflective Retrieval-Augmented Generation (Self-RAG) that enhances an LM's quality and factuality through retrieval and self-reflection. Our framework trains a single arbitrary LM that adaptively retrieves passages on-demand, and generates and reflects on retrieved passages and its own generations using special tokens, called reflection tokens. Generating reflection tokens makes the LM controllable during the inference phase, enabling it to tailor its behavior to diverse task requirements. Experiments show that Self-RAG (7B and 13B parameters) significantly outperforms state-of-the-art LLMs and retrieval-augmented models on a diverse set of tasks. Specifically, Self-RAG outperforms ChatGPT and retrieval-augmented Llama2-chat on Open-domain QA, reasoning and fact verification tasks, and it shows significant gains in improving factuality and citation accuracy for long-form generations relative to these models.

Key facts. Cited by 1,800 papers in the map (data through 7 Oct 2026). Placed in RAG robustness and adaptive retrieval, one of 15 directions in Retrieval-augmented generation. Builds on 53 of its 57 references that are in the map. First posted on arXiv 17 Oct 2023.

Builds on

53 of 57 references.

References · 53

  1. 2023GPT-4 Technical Reportcited by 18,132
  2. 2022Training language models to follow instructions with human feedbackcited by 15,712
  3. 2017Proximal Policy Optimization Algorithmscited by 14,441
  4. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
  5. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
  6. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,767
  7. 2018Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challengecited by 4,710
  8. 2023Toolformer: Language Models Can Teach Themselves to Use Toolscited by 4,072
  9. 2021Finetuned Language Models Are Zero-Shot Learnerscited by 3,747
  10. 2022FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awarenesscited by 3,611
  11. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,317
  12. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,895
38 more
  1. 2023G-Eval: NLG Evaluation using GPT-4 with Better Human Alignmentcited by 2,205
  2. 2020REALM: Retrieval-Augmented Language Model Pre-Trainingcited by 2,126
  3. 2019Fine-Tuning Language Models from Human Preferencescited by 2,124
  4. 2018Can a Suit of Armor Conduct Electricity? A New Dataset for Open Book Question Answeringcited by 2,114
  5. 2021WebGPT: Browser-assisted question-answering with human feedbackcited by 1,722
  6. 2021Multitask Prompted Training Enables Zero-Shot Task Generalizationcited by 1,462
  7. 2018FEVER: a large-scale dataset for Fact Extraction and VERificationcited by 1,362
  8. 2019ZeRO: Memory Optimizations Toward Training Trillion Parameter Modelscited by 1,278
  9. 2023FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generationcited by 1,259
  10. 2021Unsupervised Dense Information Retrieval with Contrastive Learningcited by 1,203
  11. 2022When Not to Trust Language Models: Investigating Effectiveness of Parametric and Non-Parametric Memoriescited by 1,103
  12. 2022Atlas: Few-shot Learning with Retrieval Augmented Language Modelscited by 940
  13. 2023Large Language Models Can Be Easily Distracted by Irrelevant Contextcited by 820
  14. 2019CTRL: A Conditional Transformer Language Model for Controllable Generationcited by 754
  15. 2023In-Context Retrieval-Augmented Language Modelscited by 694
  16. 2023Enabling Large Language Models to Generate Text with Citationscited by 584
  17. 2020KILT: a Benchmark for Knowledge Intensive Language Taskscited by 571
  18. 2023Language Agent Tree Search Unifies Reasoning Acting and Planning in Language Modelscited by 551
  19. 2018Wizard of Wikipedia: Knowledge-Powered Conversational agentscited by 530
  20. 2023Active Retrieval Augmented Generationcited by 506
  21. 2021MAUVE: Measuring the Gap Between Neural Text and Human Text using Divergence Frontierscited by 466
  22. 2021Large Dual Encoders Are Generalizable Retrieverscited by 454
  23. 2023How Far Can Camels Go? Exploring the State of Instruction Tuning on Open Resourcescited by 405
  24. 2023Fine-Grained Human Feedback Gives Better Rewards for Language Model Trainingcited by 399
  25. 2023Making Retrieval-Augmented Language Models Robust to Irrelevant Contextcited by 364
  26. 2023Chain-of-Verification Reduces Hallucination in Large Language Modelscited by 317
  27. 2023Evaluating Verifiability in Generative Search Enginescited by 297
  28. 2022Teaching language models to support answers with verified quotescited by 290
  29. 2023How is ChatGPT's behavior changing over time?cited by 265
  30. 2023RECOMP: Improving Retrieval-Augmented LMs with Compression and Selective Augmentationcited by 253
  31. 2023Pretraining Language Models with Human Preferencescited by 245
  32. 2022Quark: Controllable Text Generation with Reinforced Unlearningcited by 236
  33. 2022ASQA: Factoid Questions Meet Long-Form Answerscited by 231
  34. 2023REFINER: Reasoning Feedback on Intermediate Representationscited by 203
  35. 2019Learning to Retrieve Reasoning Paths over Wikipedia Graph for Question Answeringcited by 147
  36. 2022Attributed Question Answering: Evaluation and Modeling for Attributed Large Language Modelscited by 108
  37. 2022Task-aware Retrieval with Instructionscited by 107
  38. 2019A Discrete Hard EM Approach for Weakly Supervised Question Answeringcited by 78

50 most cited of 53 All 53 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 14; 2024: 380; 2025: 582; 2026: 824.

1,800 papers.

Sort:

By area:

  1. 2023Retrieval-Augmented Generation for Large Language Models: A Surveycited by 2,185
  2. 2025Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learningcited by 1,371
  3. 2025A-MEM: Agentic Memory for LLM Agentscited by 1,000
  4. 2024From Generation to Judgment: Opportunities and Challenges of LLM-as-a-judgecited by 480
  5. 2024LLMs-as-Judges: A Comprehensive Survey on LLM-based Evaluation Methodscited by 451
  6. 2024Adaptive-RAG: Learning to Adapt Retrieval-Augmented Large Language Models through Question Complexitycited by 437
  7. 2024Retrieval-Augmented Generation for AI-Generated Content: A Surveycited by 352
  8. 2025Agentic Context Engineering: Evolving Contexts for Self-Improving Language Modelscited by 349
  9. 2025R1-Searcher: Incentivizing the Search Capability in LLMs via Reinforcement Learningcited by 322
  10. 2025WebThinker: Empowering Large Reasoning Models with Deep Research Capabilitycited by 320
  11. 2025Memory in the Age of AI Agentscited by 285
  12. 2024A Survey on Knowledge Distillation of Large Language Modelscited by 276
38 more
  1. 2024The Power of Noise: Redefining Retrieval for RAG Systemscited by 275
  2. 2024RAFT: Adapting Language Model to Domain Specific RAGcited by 258
  3. 2024Personal LLM Agents: Insights and Survey about the Capability, Efficiency and Securitycited by 257
  4. 2024MinerU: An Open-Source Solution for Precise Document Content Extractioncited by 255
  5. 2024PoisonedRAG: Knowledge Corruption Attacks to Retrieval-Augmented Generation of Large Language Modelscited by 251
  6. 2024VisRAG: Vision-based Retrieval-augmented Generation on Multi-modality Documentscited by 237
  7. 2025Evaluating Memory in LLM Agents via Incremental Multi-Turn Interactionscited by 236
  8. 2024When Can LLMs Actually Correct Their Own Mistakes? A Critical Survey of Self-Correction of LLMscited by 230
  9. 2024Generative Representational Instruction Tuningcited by 227
  10. 2024AgentClinic: a multimodal agent benchmark to evaluate AI in simulated clinical environmentscited by 215
  11. 2025A Survey on Test-Time Scaling in Large Language Models: What, How, Where, and How Well?cited by 215
  12. 2025ZeroSearch: Incentivize the Search Capability of LLMs without Searchingcited by 203
  13. 2024RankRAG: Unifying Context Ranking with Retrieval-Augmented Generation in LLMscited by 200
  14. 2024A Survey on Efficient Inference for Large Language Modelscited by 197
  15. 2024Corrective Retrieval Augmented Generationcited by 187
  16. 2025BrowseComp-Plus: A More Fair and Transparent Evaluation Benchmark of Deep-Research Agentcited by 183
  17. 2024Don't Hallucinate, Abstain: Identifying LLM Knowledge Gaps via Multi-LLM Collaborationcited by 176
  18. 2024A Survey on Recent Advances in LLM-Based Multi-turn Dialogue Systemscited by 170
  19. 2024FlashRAG: A Modular Toolkit for Efficient Retrieval-Augmented Generation Researchcited by 165
  20. 2025Multi-agent Architecture Search via Agentic Supernetcited by 164
  21. 2024BRIGHT: A Realistic and Challenging Benchmark for Reasoning-Intensive Retrievalcited by 153
  22. 2025ReSearch: Learning to Reason with Search for LLMs via Reinforcement Learningcited by 152
  23. 2024Fine-grained Hallucination Detection and Editing for Language Modelscited by 146
  24. 2025Dynamic Cheatsheet: Test-Time Learning with Adaptive Memorycited by 146
  25. 2024RQ-RAG: Learning to Refine Queries for Retrieval Augmented Generationcited by 144
  26. 2024Certifiably Robust RAG against Retrieval Corruptioncited by 144
  27. 2026SimpleMem: Efficient Lifelong Memory for LLM Agentscited by 140
  28. 2024The Revolution of Multimodal Large Language Models: A Surveycited by 134
  29. 2025Reinforcing Spatial Reasoning in Vision-Language Models with Interwoven Thinking and Visual Drawingcited by 130
  30. 2025LLM Post-Training: A Deep Dive into Reasoning Large Language Modelscited by 129
  31. 2025Evo-Memory: Benchmarking LLM Agent Test-time Learning with Self-Evolving Memorycited by 129
  32. 2024GNN-RAG: Graph Neural Retrieval for Large Language Model Reasoningcited by 128
  33. 2023A Survey of Large Language Models in Medicine: Progress, Application, and Challengecited by 123
  34. 2024Long-Context LLMs Meet RAG: Overcoming Challenges for Long Inputs in RAGcited by 122
  35. 2025Deep Research Agents: A Systematic Examination And Roadmapcited by 122
  36. 2025MMSearch-R1: Incentivizing LMMs to Searchcited by 120
  37. 2024HELMET: How to Evaluate Long-Context Language Models Effectively and Thoroughlycited by 119
  38. 2024Retrieval Augmented Generation or Long-Context LLMs? A Comprehensive Study and Hybrid Approachcited by 117

50 most cited of 1,800 All 1,800 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025SKILL-RAG: Self-Knowledge Induced Learning and Filtering for Retrieval-Augmented Generation
  2. 2025Post-training an LLM for RAG? Train on Self-Generated Demonstrations
  3. 2025Retrieval is Not Enough: Enhancing RAG Reasoning through Test-Time Critique and Optimization
  4. 2024Inference Scaling for Bridging Retrieval and Augmented Generation
  5. 2026RAG-TESTER: Automated End-to-End Testing of Retrieval-Augmented Large Language Models
  6. 2024A Survey on RAG Meeting LLMs: Towards Retrieval-Augmented Large Language Models
  7. 2025When Retrieval Succeeds and Fails: Rethinking Retrieval-Augmented Generation for LLMs
  8. 2025Injecting External Knowledge into the Reasoning Process Enhances Retrieval-Augmented Generation