Adaptive Retrieval Without Self-Knowledge? Bringing Uncertainty Back Home

2025/01/22 by Viktor Moskvoretskii, Moskvoretskii, Viktor, Maria Lysyuk +17 · 4 citations
Computer Science · #Computation and Language (cs.CL) #FOS: Computer and information sciences #Information Retrieval and Search Behavior #Machine Learning (cs.LG) #Semantic Web and Ontologies

paper · pdf · doi:10.48550/arxiv.2501.12835

openalex publication_date 2025/01/22 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Retrieval Augmented Generation (RAG) improves correctness of Question Answering (QA) and addresses hallucinations in Large Language Models (LLMs), yet greatly increase computational costs. Besides, RAG is not always needed as may introduce irrelevant information. Recent adaptive retrieval methods integrate LLMs' intrinsic knowledge with external information appealing to LLM self-knowledge, but they often neglect efficiency evaluations and comparisons with uncertainty estimation techniques. We bridge this gap by conducting a comprehensive analysis of 35 adaptive retrieval methods, including 8 recent approaches and 27 uncertainty estimation techniques, across 6 datasets using 10 metrics for QA performance, self-knowledge, and efficiency. Our findings show that uncertainty estimation techniques often outperform complex pipelines in terms of efficiency and self-knowledge, while maintaining comparable QA performance.

Cited by

• Optimizing Retrieval for RAG via Reinforced Contrastive Learning
• Uncertainty Quantification for Retrieval-Augmented Reasoning
• Do LVLMs Know What They Know? A Systematic Study of Knowledge Boundary Perception in LVLMs
• Fast, Slow, and Tool-augmented Thinking for LLMs: A Review

Related

• To Retrieve or Not to Retrieve? Uncertainty Detection for Dynamic Retrieval Augmented Generation
• LLM-Independent Adaptive RAG: Let the Question Speak for Itself
• SUGAR: Leveraging Contextual Confidence for Smarter Retrieval
• SeaKR: Self-aware Knowledge Retrieval for Adaptive Retrieval Augmented Generation
• Beyond Self-Knowledge: Propagating Uncertainty Across Reasoning and Retrieval in LLMs