Evidence sources
Reasoning levels
LLM backbones
Why must retail agents bridge sources?
Enterprise retail workflows require agents to chain evidence across heterogeneous sources, from structured product and order records to unstructured inquiries and reviews.
Yet real enterprise data is rarely accessible, and existing agentic RAG benchmarks rarely capture these cross-source dependencies.
We introduce RetailBridge, a compositional benchmark built from persona-driven retail simulations, and RetailRAG, a retail-aware framework for evaluating LLM agents on it.
- 1
Identify the relevant sources
Decide which tables and text collections hold the evidence a question needs.
- 2
Carry evidence between them
Pass entities and intermediate answers from one source into the next.
- 3
Produce a grounded answer
Return an answer backed by the records that were actually retrieved.
Which LLMs can bridge evidence between sources in a heterogeneous retail database?
We evaluate 16 LLM backbones on RetailBridge using the same RetailRAG pipeline.
- Answer accuracy
- LLM-as-judge (Muse Ember)
- Source routing F1
- queried vs. gold sources
- Efficiency
- mean API calls per query
Answer accuracy
HIGHER IS BETTER ↑Loading published results…
Answer accuracy vs. cost per queryWhich backbones give the most accuracy per dollar?
UP AND LEFT IS BETTER ↖Answer accuracy: LLM-as-judge (Muse Ember), overall. Cost per query: mean USD per question on OpenRouter (log scale), covering every LLM call made while answering; it excludes the one-time schema briefing, retrieval embeddings, and judge calls.
RetailBridgeMulti-Source QA over Heterogeneous Retail Sources
Alongside RetailBridge, we propose RetailRAG as a retail-aware reference framework for evaluating LLMs over heterogeneous retail sources.
RetailRAGA Retail-Aware Agentic RAG Framework
Qwen3.6-Max
DeepSeek-V4-Pro
Gemini-3.1-Pro
GPT-5.5
Claude Opus 4.7Human annotators
Reasoning Capability
One, two, or three hops reveal where compositional reasoning breaks down.
- Level-124014.3%
- Level-272042.9%
- Level-372042.9%
Multi-Source Chaining
Single-source and multi-source queries isolate the challenge of bridging evidence.
- Homo72042.9%
- Hetero96057.1%
Evidence Modality
SQL, text, and combined evidence test both retrieval and cross-modality reasoning.
- SQL-only60035.7%
- Text-only60035.7%
- Composite48028.6%
What is the average original price of the products purchased from Robert A. by the buyer who gave a 5-star review for Robert A.'s Libby's Mixed Vegetables, mentioning their 'soup vibe had just passed'?
- 1REVIEWS_TEXT · POST-CS · 5 / 5Find the reviewerHOP 1 · TEXT RETRIEVAL
The support person was so sweet and totally understood that my soup vibe had just passed, which made the whole return feel really easy. I love that they were so kind and didn't make me feel weird about my change of heart!
BUYER IDENTIFIEDStanley M. - 2ORDERS · STRUCTURED RECORDSList the purchasesHOP 2 · SQL QUERY
SELECT DISTINCT product_name FROM "Order" WHERE seller_name = 'Robert A.' AND buyer_name = 'Stanley M.'
QUERY RESULT · 4 PRODUCTS- Libby's Mixed Vegetables …
- Jelly Belly 20-Flavor Spring Gift Box, 8.5-oz
- SUNMALL 821-2049-A HDD Cable …
- FLYRUN Men's Athletic Ankle Quarter Socks …
- 3PRODUCTS · STRUCTURED RECORDSAverage the pricesHOP 3 · SQL QUERY
SELECT AVG(original_price) FROM Product WHERE product_name IN ( 'Libby''s Mixed …', 'Jelly Belly …', 'SUNMALL …', 'FLYRUN …')
BENCHMARK GOLD ANSWER15.9
Cite This Work
@article{choi2026retailbridge,
title={Toward Agentic Retrieval-Augmented
Generation for Enterprise Retail Workflows
over Heterogeneous Data Sources},
author={Choi, Jeonghwan and Hwang, Jibin and
Park, Hayoon and Cheon, Hyeonjae and
Song, Hwanjun},
journal={arXiv preprint arXiv:2610.25096},
year={2026}
}