Skip to content

MERGE: Multi-LLM Ensemble for Retrieval via Generative Enrichment

Sep 2026 · arXiv.org · Vol abs/2609.37574 · 0 citations · 39 references
Computer Science

Abstract

Large Language Models (LLMs) are increasingly used to enrich user queries in information retrieval (IR) so that a standard retriever such as BM25 can bridge vocabulary gaps with the target corpus. Any single LLM, however, is limited by its training data and architectural biases, and its enrichment behavior depends on hand-crafted prompts that must be re-engineered for each new model -- an expensive and poorly scalable process. We present MERGE (Multi-LLM Ensemble for Retrieval via Generative Enrichment), a two-stage framework: three heterogeneous 7-8B open-source LLMs independently produce candidate expansions, and a larger LLM generatively synthesizes them into a single query. To make prompt engineering scalable across the ensemble, we integrate a task-grounded Automatic Prompt Optimization (APO) loop into both stages. Unlike APO methods that judge candidates with an LLM evaluator, our loop scores each candidate by its downstream retrieval performance and runs a small tournament between the current champion prompt and optimizer-proposed drafts, terminating once the champion survives two consecutive rounds; a history-augmented variant additionally feeds the recent tournament trajectory back to the optimizer. MERGE is retriever-agnostic and issues a single BM25 pass with no rank fusion, no supervised document expansion, and no re-indexing. On five BEIR benchmarks (NQ, SciFact, FiQA, Touche-2020, DBPedia), MERGE improves BM25 nDCG@10 over the original queries by +2.1 to +14.9 points and matches or outperforms strong LLM-based query-expansion baselines despite using only compact open-source models. Ablations confirm that the Stage-2 ensemble beats any single Stage-1 LLM, and that task-grounded APO converts large seed-prompt regressions into consistent gains without hand-tuning.

View source

Similar papers

#artificial intelligence Open access Sep 2026

CompCQR: Compositional Query Generation for Training-Free Conversational Search

This work proposes a novel training-free method that generates a very large number of queries with minimal LLM usage by compositionally combining a small set of atomic components and applies LLM reasoning to construct a high-quality document set that balances precision and recall while capturing the user's core intent.

Yunah Jang, Kang-il Lee, Joongbo Shin et al. · 0 citations
#small language model Preprint Aug 2026

Query Expansion Is More Than Generation: Improving Dense Retrieval through Better Integration

This work introduces AnchorQE, a training-free method that separately encodes the original query and its expansion before interpolating them, and shows that AnchorQE improves retrieval effectiveness by up to 12.89% when compared to widely-used expansion-only or text-level concatenation baselines across TREC-DL, LoTTE,...

Sixia Sun, Mihai Surdeanu · 1 citation
#artificial intelligence Preprint Sep 2026

Re-ranking and Late Interaction Drive Retrieval Quality: A Controlled Comparison of RAG Strategies for Scientific Question Answering

Retrieval-Augmented Generation (RAG) is now the standard way to ground Large Language Models (LLMs) in external knowledge, yet the design space of retrieval pipelines is large and the trade-offs between variants are not well understood, especially on domain-specific corpora at realistic scale. In this work, we present...

B. Rathi, Eshan Chawla, William B. Andreopoulos · 0 citations
Preprint Aug 2026

GEM: A Generative Embedding Model Bridging Reasoning and Retrieval

GEM is presented, a generative embedding model that augments retrieval through its own knowledge by explicitly reasoning about user intent and relevance criteria, and its generative nature allows test-time compute scaling via prompting to further enhance retrieval performance.

Zhili Shen, Craig Macdonald · 0 citations
Book Open access Sep 2026

Seek: Self-Evaluative Exploration for Knowledge Retrieval

seek, Self-Evaluative Exploration for Knowledge Retrieval, a training-free framework that addresses this limitation through iterative corpus interaction at test time through iterative corpus interaction at test time.

Amin Bigdeli, Radin Hamidi Rad, Negar Arabzadeh et al. · 1 citation

Related blog posts

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.