mlsanityData to 7 Oct 2026

RoseRAG: Robust Retrieval-augmented Generation with Small-scale LLMs via Margin-aware Preference Optimization

arXiv 2502.10993 ↗First posted 16 Feb 2025cs.CL cs.LGcited by 9

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

also in 1 more
Abstract (arXiv)

Large language models (LLMs) have achieved impressive performance but face high computational costs and latency, limiting their deployment in resource-constrained settings. In contrast, small-scale LLMs (SLMs) are more efficient yet struggle to capture evolving real-world knowledge. Retrieval-augmented generation (RAG) helps by integrating external knowledge, but imperfect retrieval can introduce distracting noise that misleads SLMs. We propose RoseRAG, a robust RAG framework for SLMs via Margin-aware Preference Optimization. RoseRAG employs multi-turn prompting for detailed reasoning, rejection sampling for high-quality explanations, and contrastive preference selection to refine responses by maximizing the likelihood gap between preferred and non-preferred outputs. By integrating these components into a margin-aware optimization process, RoseRAG robustly enhances the accuracy and reliability of SLMs for RAG applications. Extensive experiments on three open-domain question answering benchmarks indicate that our innovative RoseRAG surpasses state-of-the-art baselines significantly.

Key facts. Cited by 9 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 55 of its 67 references that are in the map. First posted on arXiv 16 Feb 2025.

Builds on

55 of 67 references.

Closest ancestors · 7

Same direction, newest first

  1. 2024SimRAG: Self-Improving Retrieval-Augmented Generation for Adapting Large Language Models to Specialized Domains23 Oct
  2. 2024KnowPO: Knowledge-aware Preference Optimization for Controllable Knowledge Selection in Retrieval-Augmented Language Models6 Aug
  3. 2024InstructRAG: Instructing Retrieval-Augmented Generation via Self-Synthesized Rationales19 Jun
  4. 2024Enhancing Noise Robustness of Retrieval-Augmented Language Models with Adaptive Adversarial Training31 May
  5. 2024Adaptive-RAG: Learning to Adapt Retrieval-Augmented Large Language Models through Question Complexity21 Mar
  6. 2024BlendFilter: Advancing Retrieval-Augmented Large Language Models via Query Generation Blending and Knowledge Filtering16 Feb
  7. 2023Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection17 Oct

Other references · 48

  1. 2023GPT-4 Technical Reportcited by 18,132
  2. 2022Training language models to follow instructions with human feedbackcited by 15,712
  3. 2024The Llama 3 Herd of Modelscited by 14,192
  4. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  5. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
  6. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,393
  7. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,355
  8. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,767
  9. 2019Parameter-Efficient Transfer Learning for NLPcited by 4,195
  10. 2021Prefix-Tuning: Optimizing Continuous Prompts for Generationcited by 3,878
  11. 2021The Power of Scale for Parameter-Efficient Prompt Tuningcited by 3,769
  12. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,686
36 more
  1. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,634
  2. 2020Learning to summarize from human feedbackcited by 2,562
  3. 2024Gemma 2: Improving Open Language Models at a Practical Sizecited by 1,928
  4. 2020Constructing A Multi-hop QA Dataset for Comprehensive Evaluation of Reasoning Stepscited by 1,259
  5. 2022Measuring and Narrowing the Compositionality Gap in Language Modelscited by 1,054
  6. 2024SimPO: Simple Preference Optimization with a Reference-Free Rewardcited by 951
  7. 2022Atlas: Few-shot Learning with Retrieval Augmented Language Modelscited by 940
  8. 2024Length-Controlled AlpacaEval: A Simple Way to Debias Automatic Evaluatorscited by 865
  9. 2020What Makes for Good Views for Contrastive Learning?cited by 843
  10. 2023DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelinescited by 803
  11. 2023REPLUG: Retrieval-Augmented Black-Box Language Modelscited by 786
  12. 2023AlpacaFarm: A Simulation Framework for Methods that Learn from Human Feedbackcited by 738
  13. 2021Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit Reasoning Strategiescited by 736
  14. 2025SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-trainingcited by 655
  15. 2020Prototypical Contrastive Learning of Unsupervised Representationscited by 636
  16. 2020Contrastive Learning with Hard Negative Samplescited by 619
  17. 2024ORPO: Monolithic Preference Optimization without Reference Modelcited by 556
  18. 2024Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Modelscited by 552
  19. 2021ColBERTv2: Effective and Efficient Retrieval via Lightweight Late Interactioncited by 464
  20. 2024Contrastive Preference Optimization: Pushing the Boundaries of LLM Performance in Machine Translationcited by 390
  21. 2023Making Retrieval-Augmented Language Models Robust to Irrelevant Contextcited by 364
  22. 2023Enhancing Retrieval-Augmented Large Language Models with Iterative Retrieval-Generation Synergycited by 343
  23. 2023Statistical Rejection Sampling Improves Preference Optimizationcited by 321
  24. 2022Demonstrate-Search-Predict: Composing retrieval and language models for knowledge-intensive NLPcited by 316
  25. 2021Parametric Contrastive Learningcited by 217
  26. 2024RankRAG: Unifying Context Ranking with Retrieval-Augmented Generation in LLMscited by 200
  27. 2024ReFT: Representation Finetuning for Language Modelscited by 186
  28. 2023Chain-of-Note: Enhancing Robustness in Retrieval-Augmented Language Modelscited by 180
  29. 2023Fine-Tuning or Retrieval? Comparing Knowledge Injection in LLMscited by 175
  30. 2024Certifiably Robust RAG against Retrieval Corruptioncited by 144
  31. 2024Small Language Models: Survey, Measurements, and Insightscited by 109
  32. 2023Answering Questions by Meta-Reasoning over Multiple Chains of Thoughtcited by 87
  33. 2023LLatrieval: LLM-Verified Retrieval for Verifiable Generationcited by 52
  34. 2023Small Language Models Improve Giants by Rewriting Their Outputscited by 18
  35. 2024Large Convolutional Model Tuning via Filter Subspacecited by 11
  36. 2024Enhancing Robustness of Retrieval-Augmented Language Models with In-Context Learningcited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 6; 2026: 3.

9 papers.

  1. 2025Collab-RAG: Boosting Retrieval-Augmented Generation for Complex Question Answering via White-Box and Black-Box LLM Collaborationcited by 19
  2. 2025Unlocking Efficient, Scalable, and Continual Knowledge Editing with Basis-Level Representation Fine-Tuningcited by 17
  3. 2025Retrieval-augmented GUI Agents with Generative Guidelinescited by 14
  4. 2025AceSearcher: Bootstrapping Reasoning and Search for LLMs via Reinforced Self-Playcited by 10
  5. 2025Curriculum Guided Reinforcement Learning for Efficient Multi Hop Retrieval Augmented Generationcited by 5
  6. 2025In-Context Compositional Learning via Sparse Coding Transformercited by 0
  7. 2026From 50% to Mastery in 3 Days: A Low-Resource SOP for Localizing Graduate-Level AI Tutors via Shadow-RAGcited by 0
  8. 2026LegalDrill: Diagnosis-Driven Synthesis for Legal Reasoning in Small Language Modelscited by 0
  9. 2026RIMS: Preference Optimization via Smoothed Multi-pair Aggregation for Small-Scale LLM Retrieval-Augmented Generationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Retrieval-Augmented Generation: A Comprehensive Survey of Architectures, Enhancements, and Robustness Frontiers
  2. 2025Investigating the Robustness of Retrieval-Augmented Generation at the Query Level
  3. 2025Optimizing Knowledge Integration in Retrieval-Augmented Generation with Self-Selection
  4. 2025GainRAG: Preference Alignment in Retrieval-Augmented Generation through Gain Signal Synthesis
  5. 2024Bridging the Preference Gap between Retrievers and LLMs
  6. 2024Understand What LLM Needs: Dual Preference Alignment for Retrieval-Augmented Generation
  7. 2025VaccineRAG: Boosting Multimodal Large Language Models' Immunity to Harmful RAG Samples
  8. 2025Adaptive Retrieval Without Self-Knowledge? Bringing Uncertainty Back Home