# Knowing You Don't Know: Learning When to Continue Search in Multi-round RAG through Self-Practicing

arXiv: https://arxiv.org/abs/2505.02811 · HTML: https://mlsanity.com/paper/2505.02811/

## Key facts

- Cited by 25 papers in the map (data through 14 Oct 2026).
- Placed in Conversational and speech RAG, one of 15 directions in Retrieval-augmented generation.
- Builds on 34 of its 46 references that are in the map.
- First posted on arXiv 5 May 2025.

## Placed in

- Retrieval-augmented generation / [Conversational and speech RAG](https://mlsanity.com/rag/conversational-and-speech-rag/)

## Builds on (most cited first, 30 of 34)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Let's Verify Step by Step](https://mlsanity.com/paper/2305.20050/) (2023)
- [HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering](https://mlsanity.com/paper/1809.09600/) (2018)
- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension](https://mlsanity.com/paper/1705.03551/) (2017)
- [Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection](https://mlsanity.com/paper/2310.11511/) (2023)
- [Improving language models by retrieving from trillions of tokens](https://mlsanity.com/paper/2112.04426/) (2021)
- [Constructing A Multi-hop QA Dataset for Comprehensive Evaluation of Reasoning Steps](https://mlsanity.com/paper/2011.01060/) (2020)
- [Text Embeddings by Weakly-Supervised Contrastive Pre-training](https://mlsanity.com/paper/2212.03533/) (2022)
- [Large Language Models Cannot Self-Correct Reasoning Yet](https://mlsanity.com/paper/2310.01798/) (2023)
- [Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMs](https://mlsanity.com/paper/2306.13063/) (2023)
- [Atlas: Few-shot Learning with Retrieval Augmented Language Models](https://mlsanity.com/paper/2208.03299/) (2022)
- [STaR: Bootstrapping Reasoning With Reasoning](https://mlsanity.com/paper/2203.14465/) (2022)
- [Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questions](https://mlsanity.com/paper/2212.10509/) (2022)
- [Large Language Models Can Be Easily Distracted by Irrelevant Context](https://mlsanity.com/paper/2302.00093/) (2023)
- [CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquing](https://mlsanity.com/paper/2305.11738/) (2023)
- [In-Context Retrieval-Augmented Language Models](https://mlsanity.com/paper/2302.00083/) (2023)
- [Adaptive-RAG: Learning to Adapt Retrieval-Augmented Large Language Models through Question Complexity](https://mlsanity.com/paper/2403.14403/) (2024)
- [Training Language Models to Self-Correct via Reinforcement Learning](https://mlsanity.com/paper/2409.12917/) (2024)
- [Beyond Human Data: Scaling Self-Training for Problem-Solving with Language Models](https://mlsanity.com/paper/2312.06585/) (2023)
- [MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries](https://mlsanity.com/paper/2401.15391/) (2024)
- [How Context Affects Language Models' Factual Predictions](https://mlsanity.com/paper/2005.04611/) (2020)
- [FlashRAG: A Modular Toolkit for Efficient Retrieval-Augmented Generation Research](https://mlsanity.com/paper/2405.13576/) (2024)
- [On the Self-Verification Limitations of Large Language Models on Reasoning and Planning Tasks](https://mlsanity.com/paper/2402.08115/) (2024)
- [Self-Knowledge Guided Retrieval Augmentation for Large Language Models](https://mlsanity.com/paper/2310.05002/) (2023)
- [IM-RAG: Multi-Round Retrieval-Augmented Generation Through Learning Inner Monologues](https://mlsanity.com/paper/2405.13021/) (2024)
- [Worse than Zero-shot? A Fact-Checking Dataset for Evaluating the Robustness of RAG Against Misleading Retrievals](https://mlsanity.com/paper/2502.16101/) (2025)

## Built on by (most cited first, 25 of 25)

- [A Systematic Review of Key Retrieval-Augmented Generation (RAG) Systems: Progress, Gaps, and Future Directions](https://mlsanity.com/paper/2507.18910/) (2025)
- [COMI: Coarse-to-fine Context Compression via Marginal Information Gain](https://mlsanity.com/paper/2602.01719/) (2026)
- [QuCo-RAG: Quantifying Uncertainty from the Pre-training Corpus for Dynamic Retrieval-Augmented Generation](https://mlsanity.com/paper/2512.19134/) (2025)
- [S2G-RAG: Structured Sufficiency and Gap Judging for Iterative Retrieval-Augmented QA](https://mlsanity.com/paper/2604.23783/) (2026)
- [Scaling Towards the Information Boundary of Instruction Sets: The Infinity Instruct Subject Technical Report](https://mlsanity.com/paper/2507.06968/) (2025)
- [Towards Transparent RAG: Fostering Evidence Traceability in LLM Generation via Reinforcement Learning](https://mlsanity.com/paper/2505.13258/) (2025)
- [Test-time Corpus Feedback: From Retrieval to RAG](https://mlsanity.com/paper/2508.15437/) (2025)
- [How You Ask Matters! Adaptive RAG Robustness to Query Variations](https://mlsanity.com/paper/2604.10745/) (2026)
- [HALT: Verification-Aware Stopping for Retrieval-Augmented Search Agents](https://mlsanity.com/paper/2608.02009/) (2026)
- [One Patient, Many Contexts: Scaling Medical AI with Contextual Intelligence](https://mlsanity.com/paper/2506.10157/) (2025)
- [Knowledge Is Not Static: Order-Aware Hypergraph RAG for Language Models](https://mlsanity.com/paper/2604.12185/) (2026)
- [EtCon: Edit-then-Consolidate for Reliable Knowledge Editing](https://mlsanity.com/paper/2512.04753/) (2025)
- [When Iterative RAG Beats Ideal Evidence: A Diagnostic Study in Scientific Multi-hop Question Answering](https://mlsanity.com/paper/2601.19827/) (2026)
- [Beyond Relevance: On the Relationship Between Retrieval and RAG Information Coverage](https://mlsanity.com/paper/2603.08819/) (2026)
- [From Facts to Insights: A Persona-Driven Dual Memory Framework and Dataset for Role-Playing Agents](https://mlsanity.com/paper/2605.25693/) (2026)
- [Predicting Partial Answer Quality and Utility in Agentic Retrieval-Augmented Generation](https://mlsanity.com/paper/2609.16453/) (2026)
- [CoVeR: Coverage-Based Routing of Verifier Calls in Agentic Retrieval](https://mlsanity.com/paper/2609.26086/) (2026)
- [Self-MedRAG: a Self-Reflective Hybrid Retrieval-Augmented Generation Framework for Reliable Medical Question Answering](https://mlsanity.com/paper/2601.04531/) (2026)
- [QChunker: Learning Question-Aware Text Chunking for Domain RAG via Multi-Agent Debate](https://mlsanity.com/paper/2603.11650/) (2026)
- [A Reproducibility Study of Metacognitive Retrieval-Augmented Generation](https://mlsanity.com/paper/2604.19899/) (2026)
- [When Should Multi-Round RAG Stop? Structured Stopping Judgments and Retrieval Reduction in Search-R1](https://mlsanity.com/paper/2608.13237/) (2026)
- [SurveyAgent-HKA: A multi-agent framework for scientific survey generation with LLMs and human knowledge augmentation](https://mlsanity.com/paper/2609.05938/) (2026)
- [Evaluating and Improving Evidence-Grounded Fact-Checking in LLMs via Multi-Round Evidence Ablation](https://mlsanity.com/paper/2609.08943/) (2026)
- [Relevance Is Not Sufficient Evidence: Detecting Evidence Gaps Before Generation in RAG](https://mlsanity.com/paper/2609.37469/) (2026)
- [CLIMB: Confidence-Guided Complementary Evidence for Multimodal Retrieval-Augmented Generation](https://mlsanity.com/paper/2610.03421/) (2026)

## Abstract (arXiv)

Retrieval Augmented Generation (RAG) has shown strong capability in enhancing language models' knowledge and reducing AI generative hallucinations, driving its widespread use. However, complex tasks requiring multi-round retrieval remain challenging, and early attempts tend to be overly optimistic without a good sense of self-skepticism. Current multi-round RAG systems may continue searching even when enough information has already been retrieved, or they may provide incorrect answers without having sufficient information or knowledge. Existing solutions either require large amounts of expensive human-labeled process supervision data or lead to subpar performance. This paper aims to address these limitations by introducing a new framework, SIM-RAG, to explicitly enhance RAG systems' self-awareness and multi-round retrieval capabilities. To train SIM-RAG, we first let a RAG system self-practice multi-round retrieval, augmenting existing question-answer pairs with intermediate inner monologue reasoning steps to generate synthetic training data. For each pair, the system may explore multiple retrieval paths, which are labeled as successful if they reach the correct answer and unsuccessful otherwise. Using this data, we train a lightweight information sufficiency Critic. At inference time, the Critic evaluates whether the RAG system has retrieved sufficient information at each round, guiding retrieval decisions and improving system-level self-awareness through in-context reinforcement learning. Experiments across multiple prominent RAG benchmarks show that SIM-RAG is an effective multi-round RAG solution. Furthermore, this framework is system-efficient, adding a lightweight component to RAG without requiring modifications to existing LLMs or search engines, and data-efficient, eliminating the need for costly human-annotated mid-step retrieval process supervision data.