A Survey on Knowledge-Oriented Retrieval-Augmented Generation
Source: https://arxiv.org/abs/2503.10677 · Published 2025-03-11 · Authors: Mingyue Cheng, Yucong Luo, Jie Ouyang, Qi Liu, Huijie Liu, Li Li, et al. (12 authors)
Abstract (from the paper)
> Retrieval-Augmented Generation (RAG) has gained significant attention in recent years for its potential to enhance natural language understanding and generation by combining large-scale retrieval systems with generative models. RAG leverages external knowledge sources, such as documents, databases, or structured data, to improve model performance and generate more accurate and contextually relevant outputs. This survey provides a comprehensive overview of RAG by examining its fundamental components, including retrieval mechanisms, generation processes, and the integration between the two. It presents a taxonomy categorizing RAG methods, from basic retrieval-augmented approaches to advanced models incorporating multi-modal data and reasoning capabilities, and reviews evaluation benchmarks, datasets, and applications in question answering, summarization, and information retrieval.
Key points
- Unified perspective: Consolidates scattered RAG research into a comparable framework, decomposing method components (representation learning, retrievers, rerankers, planners, generators, feedback mechanisms).
- Taxonomy dimensions:
- Four technical lines compared: dense retrieval (high recall, low latency), late interaction such as ColBERT (higher precision, larger index), generative IR (directly "generates" documents as tokens/docids), and agentic search (search as sequential decision-making with multi-hop reasoning and self-reflection).
- Timeline: 2019–2021: BERT rerankers and DPR established neural retrieval; 2022–2023: RAG and FreshLLM drove retrieval-generation fusion; 2024+: conversational/agentic search and Gen-RecSys; 2025–2026: RL-trained search agents, Deep Research, and GraphRAG.
- Evaluation benchmarks covered: MS MARCO, BEIR, Natural Questions; metrics such as nDCG@10, MRR, Recall@k, task success rate, latency and token cost.
- Lack of unified benchmarks; irreproducible private data
- LLM evaluation bias; safety and cost constraints of agentic systems
- Finer-grained process supervision; joint retrieval–reasoning training
- Enterprise metadata governance; multi-modal and cross-lingual consistency
- Deeper integration with knowledge graphs and structured databases; efficient test-time compute allocation
- A Comprehensive Survey on Reinforcement Learning-based Agentic Search (arXiv:2510.16724)
- A Survey of Conversational Search (arXiv:2410.15576)
- A Survey of LLM Empowered Agents for Recommendation (arXiv:2503.05659)
- A Survey of Model Architectures in Information Retrieval (arXiv:2502.14822)
| Dimension | Subtypes | Representative ideas | Strengths | Limitations | |---|---|---|---|---| | Modeling paradigm | Discriminative / generative retrieval | Bi-encoder, cross-encoder, DSI | Mature, scalable | Semantic drift, update cost | | LLM integration | RAG / Agent / Tool-use | Retrieval augmentation, search agents, API calls | Flexible, interpretable | Latency, error propagation | | Optimization goals | Relevance / diversity / freshness | Multi-objective LTR, RLHF, online learning | Business-aligned | Scarce annotations | | Evaluation | Offline / online / human | nDCG, MRR, LLM-as-judge, A/B | Comparable | Gap vs. real satisfaction |
Insights for search, recommendation, and personalization
1. Architecture: Cascaded retrieval + reranking + generation remains mainstream, but agentic paradigms make retrieval strategy itself learnable. 2. Data: High-quality instruction data and click/session logs are both critical; synthetic data risks knowledge leakage and distribution shift. 3. Evaluation: The gap between offline metrics and online satisfaction is widening; LLM-as-judge needs cross-validation with human evaluation. 4. Deployment: Latency, cost, interpretability, and safety are hard industrial constraints — do not optimize academic benchmarks alone.
Open problems and future directions
Related entries
Glossary
| Term | Meaning | |---|---| | RAG | Retrieval-Augmented Generation | | LTR | Learning to Rank | | nDCG | Normalized Discounted Cumulative Gain | | Agentic Search | Search modeled as sequential decision-making with tool calls | | Gen-IR | Generative Information Retrieval |