A Survey of Conversational Search
Source: arXiv:2410.15576 · Type: Survey · Published: 2024-10-21 Authors: Fengran Mo, Kelong Mao, Ziliang Zhao, Hongjin Qian, Haonan Chen, Yiruo Cheng, et al. (10 authors)
Abstract (original)
> As a cornerstone of modern information access, search engines have become indispensable in everyday life. With the rapid advancements in AI and natural language processing (NLP) technologies, particularly large language models (LLMs), search engines have evolved to support more intuitive and intelligent interactions between users and systems. Conversational search, an emerging paradigm for next-generation search engines, leverages natural language dialogue to facilitate complex and precise information retrieval, thus attracting significant attention. Unlike traditional keyword-based search engines, conversational search systems enhance user experience by supporting intricate queries, maintaining context over multi-turn interactions, and providing robust information integration and processing capabilities. Key components such as query reformulation, search clarification, conversational retrieval, and response generation work in unison to enable these sophisticated interactions. In this survey, we explore the recent advancements and potential future directions in conversational search, examining the critical modules that constitute a conversational search system. We highlight the integration of LLMs in enhancing these systems and discuss the challenges and opportunities that lie ahead in this dynamic field. Additionally, we provide insights into real-world applications and robust evaluations of current conversational search systems, aiming to guide future research and development in conversational search.
Key points
- Unified perspective: The survey consolidates dispersed work on conversational search into a comparable framework covering query reformulation, search clarification, conversational retrieval, and response generation.
- LLM integration: It highlights how LLMs enhance each module, including RAG, tool use, and agentic search patterns.
- Taxonomy of approaches:
- Four main research lines: dense retrieval (high recall, low latency), late interaction (e.g., ColBERT — higher precision, larger index), generative IR (generating docids/tokens directly), and agentic search (search as sequential decision-making with multi-hop reasoning and self-reflection).
- Evaluation practice: Common benchmarks include MS MARCO, BEIR, and Natural Questions; metrics include nDCG@10, MRR, Recall@k, task success rate, latency, and token cost; baselines span BM25, dense retrieval, cross-encoder rerankers, and retrieval-free LLMs.
- 2019–2021: BERT reranking and DPR establish neural retrieval foundations.
- 2022–2023: RAG and FreshLLMS drive retrieval–generation fusion.
- 2024 onward: Conversational/agentic search and generative recommenders (Gen-RecSys) expand rapidly.
- 2025–2026 (outlook): RL-trained search agents, Deep Research, and GraphRAG emerge as growth frontiers.
- Lack of unified benchmarks and reproducibility of private data
- Bias in LLM-based evaluation
- Safety and cost constraints for agentic systems on the open web
- Finer-grained process supervision, retrieval–reasoning co-training, and multimodal/cross-lingual consistency
- A Comprehensive Survey on RL-based Agentic Search
- LLM-Empowered Agents for Recommendation (arXiv:2503.05659)
- Model Architectures in Information Retrieval (arXiv:2502.14822)
- Knowledge-Oriented Retrieval-Augmented Generation (arXiv:2503.10677)
| Dimension | Subtypes | Representative ideas | Strengths | Limitations | |---|---|---|---|---| | Modeling paradigm | Discriminative / generative retrieval | Bi-encoders, cross-encoders, 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 testing | Comparable | Gap with real satisfaction |
Timeline of the field
Conclusions and insights
1. Architecture: Cascaded retrieve–rerank–generate remains mainstream, but the agentic paradigm makes 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 requires cross-validation with human assessment. 4. Product constraints: Latency, cost, interpretability, and safety are hard constraints for real-world deployment—optimizing academic benchmarks alone is insufficient.