English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

Hypothetical Documents or Knowledge Leakage? Rethinking LLM-based Query Expansion (arXiv, Apr 2025)

Forum topic · 小凯 · 2026-07-05

Summary

This post discusses the April 2025 arXiv paper 'Hypothetical Documents or Knowledge Leakage? Rethinking LLM-based Query Expansion' by Yejun Yoon, Jaeyoon Jung, Seunghyun Yoon, and Kunwoo Park (arXiv:2504.14175). LLM-based query expansion techniques such as HyDE generate hypothetical documents from a query and use their embeddings to improve retrieval. The paper questions whether reported gains truly come from better intent understanding, or whether they stem from knowledge leakage: when expansion relies on parametric knowledge already memorized by the LLM, benchmarks can be inflated because the model reconstructs the very target documents it was evaluated against. The forum entry summarizes the motivation, positions the work within modern information retrieval research (dense retrieval, reranking, RAG, agentic search), outlines typical evaluation protocols (MS MARCO, BEIR, nDCG@10, MRR), and derives engineering takeaways on cost, latency, evaluation credibility, and safe deployment of LLM-based expansion in production search and recommendation systems.

Hypothetical Documents or Knowledge Leakage? Rethinking LLM-based Query Expansion (arXiv, Apr 2025)

This entry summarizes the arXiv paper "Hypothetical Documents or Knowledge Leakage? Rethinking LLM-based Query Expansion" (arXiv:2504.14175), authored by Yejun Yoon, Jaeyoon Jung, Seunghyun Yoon, and Kunwoo Park, filed under the Query Understanding section of the source list.

Key points

  • Topic: LLM-based query expansion, most prominently the HyDE (Hypothetical Document Embeddings) family of methods, where an LLM generates a hypothetical answer document for a query, and retrieval is performed by embedding that generated document instead of the raw query.
  • Central question: Do the observed retrieval improvements reflect genuine understanding of user intent, or do they result from knowledge leakage — i.e., the LLM's parametric memory already contains the benchmark's target documents, allowing it to effectively reconstruct the answer during expansion?
  • Implication: If leakage is the dominant mechanism, reported benchmark gains overstate how well these methods will generalize to private corpora, fresh data, or domains unseen during the LLM's pretraining.
  • Context within IR research

    The paper sits at the intersection of neural information retrieval and LLM-era system design:

    1. Representation: queries and documents encoded as dense/sparse vectors, or hypothetical documents produced by generative models; 2. Pipeline: retriever → reranker → generator, with agentic paradigms increasingly making the number and strategy of retrieval steps learnable; 3. Evaluation: standard setups use benchmarks such as MS MARCO and BEIR with metrics like nDCG@10, MRR, and Recall@k — precisely the settings most susceptible to knowledge leakage effects.

    Takeaways for practitioners

  • Architecture: cascade retrieval + reranking + generation remains mainstream, but synthetic expansion data must be checked for leakage and distribution shift.
  • Evaluation: offline metric gains should be cross-validated with human assessment and, ideally, fresh or non-public corpora to rule out memorization effects.
  • Production constraints: latency, token cost, and safety filtering remain hard requirements; academic benchmarks alone do not justify deployment.
  • Limitations and future work

    Typical open questions in this line of research include evaluation credibility under memorization, latency and cost trade-offs, hallucination risks in generated pseudo-documents, and cross-lingual generalization. Readers should consult the original PDF for exact experimental numbers before citing quantitative results.

    Related entries

  • Aligned Query Expansion: Efficient Query Expansion for Information Retrieval (arXiv:2507.11042)
  • Beyond the Limitation of a Single Query: Train Your LLM for Query Expansion (arXiv:2510.10009)
  • Decomposing Complex Queries for Tip-of-the-tongue Retrieval (arXiv:2305.15053)
  • Few-Shot Generative Conversational Query Rewriting (SIGIR 2020, DOI: 10.1145/3397271.3401323)
  • LLM-Based Query Expansion with Gaussian Kernel Semantic Enhancement
Source: https://arxiv.org/abs/2504.14175

Tags

#information-retrieval#llm#query-expansion#hyde#knowledge-leakage#rag#dense-retrieval#arxiv

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178208802