The Attribution Crisis in LLM Search Results
Not on the map. Why?
Abstract (arXiv)
Web-enabled LLMs frequently answer queries without crediting the web pages they consume, creating an "attribution gap" - the difference between relevant URLs read and those actually cited. Drawing on approximately 14,000 real-world LMArena conversation logs with search-enabled LLM systems, we document three exploitation patterns: 1) No Search: 34% of Google Gemini and 24% of OpenAI GPT-4o responses are generated without explicitly fetching any online content; 2) No citation: Gemini provides no clickable citation source in 92% of answers; 3) High-volume, low-credit: Perplexity's Sonar visits approximately 10 relevant pages per query but cites only three to four. A negative binomial hurdle model shows that the average query answered by Gemini or Sonar leaves about 3 relevant websites uncited, whereas GPT-4o's tiny uncited gap is best explained by its selective log disclosures rather than by better attribution. Citation efficiency - extra citations provided per additional relevant web page visited - varies widely across models, from 0.19 to 0.45 on identical queries, underscoring that retrieval design, not technical limits, shapes ecosystem impact. We recommend a transparent LLM search architecture based on standardized telemetry and full disclosure of search traces and citation logs.
Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 13 of its 34 references that are in the map. First posted on arXiv 27 Jun 2025.
Builds on
13 of 34 references.
References · 13
- 2023Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflectioncited by 1,839
- 2019Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward Responsible AIcited by 1,552
- 2021Improving language models by retrieving from trillions of tokenscited by 1,293
- 2023Enabling Large Language Models to Generate Text with Citationscited by 595
- 2023Automatic Evaluation of Attribution by Large Language Modelscited by 72
- 2023A Survey of Large Language Models Attributioncited by 71
- 2024Authorship Attribution in the Era of LLMs: Problems, Methodologies, and Challengescited by 47
- 2024Source-Aware Training Enables Knowledge Attribution in Language Modelscited by 30
- 2023Citation: A Key to Building Responsible and Accountable Large Language Modelscited by 20
- 2025Leveraging Explainable AI for LLM Text Attribution: Differentiating Human-Written and Multiple LLMs-Generated Textcited by 11
- 2024Abstention vs. Hallucination: Benchmarking LLM Source Attribution for Scientific Citationscited by 10
- 2025Beyond Public Access in LLM Pre-Training Datacited by 3
Built on by
Citing papers per year: 2025: 1; 2026: 2.
3 papers.