@inproceedings{itai-etal-2026-lit,
title = "{LIT}-{RAGB}ench: Benchmarking Generator Capabilities of Large Language Models in Retrieval-Augmented Generation",
author = "Itai, Koki and
Hasegawa, Shunichi and
Yamamoto, Yuta and
Minegishi, Gouki and
Otsuki, Masaki",
editor = "Piperidis, Stelios and
Bel, N{\'u}ria and
van den Heuvel, Henk and
Ide, Nancy and
Krek, Simon and
Toral, Antonio",
booktitle = "Proceedings of the Fifteenth Language Resources and Evaluation Conference",
month = may,
year = "2026",
address = "Palma de Mallorca, Spain",
publisher = "ELRA Language Resource Association",
url = "https://aclanthology.org/2026.lrec-1.427/",
doi = "10.63317/3ep7jffajytv",
pages = "5482--5493",
abstract = "Retrieval-Augmented Generation (RAG) is a framework in which a Generator, such as a Large Language Model (LLM), produces answers by retrieving documents from an external collection using a Retriever. In practice, Generators must integrate evidence from long contexts, perform multi-step reasoning, interpret tables, and abstain when evidence is missing. However, existing benchmarks for Generators provide limited coverage, with none enabling simultaneous evaluation of multiple capabilities under unified conditions. To bridge the gap between existing evaluations and practical use, we introduce LIT-RAGBench (the Logic, Integration, Table, Reasoning, and Abstention RAG Generator Benchmark), which defines five categories: Integration, Reasoning, Logic, Table, and Abstention{---}each further divided into practical evaluation aspects. LIT-RAGBench systematically covers patterns combining multiple aspects across categories. By using fictional entities and scenarios, LIT-RAGBench evaluates answers grounded in the provided external documents. The dataset consists of 114 human-constructed Japanese questions and an English version generated by machine translation with human curation. We use LLM-as-a-Judge for scoring and report category-wise and overall accuracy. Across API-based and open-weight models, no model exceeds 90{\%} overall accuracy. By making strengths and weaknesses measurable within each category, LIT-RAGBench serves as a valuable metric for model selection in practical RAG deployments and for building RAG-specialized models."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="itai-etal-2026-lit">
<titleInfo>
<title>LIT-RAGBench: Benchmarking Generator Capabilities of Large Language Models in Retrieval-Augmented Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Koki</namePart>
<namePart type="family">Itai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shunichi</namePart>
<namePart type="family">Hasegawa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuta</namePart>
<namePart type="family">Yamamoto</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gouki</namePart>
<namePart type="family">Minegishi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Masaki</namePart>
<namePart type="family">Otsuki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Fifteenth Language Resources and Evaluation Conference</title>
</titleInfo>
<name type="personal">
<namePart type="given">Stelios</namePart>
<namePart type="family">Piperidis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Núria</namePart>
<namePart type="family">Bel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Henk</namePart>
<namePart type="family">van den Heuvel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nancy</namePart>
<namePart type="family">Ide</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Simon</namePart>
<namePart type="family">Krek</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Antonio</namePart>
<namePart type="family">Toral</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>ELRA Language Resource Association</publisher>
<place>
<placeTerm type="text">Palma de Mallorca, Spain</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Retrieval-Augmented Generation (RAG) is a framework in which a Generator, such as a Large Language Model (LLM), produces answers by retrieving documents from an external collection using a Retriever. In practice, Generators must integrate evidence from long contexts, perform multi-step reasoning, interpret tables, and abstain when evidence is missing. However, existing benchmarks for Generators provide limited coverage, with none enabling simultaneous evaluation of multiple capabilities under unified conditions. To bridge the gap between existing evaluations and practical use, we introduce LIT-RAGBench (the Logic, Integration, Table, Reasoning, and Abstention RAG Generator Benchmark), which defines five categories: Integration, Reasoning, Logic, Table, and Abstention—each further divided into practical evaluation aspects. LIT-RAGBench systematically covers patterns combining multiple aspects across categories. By using fictional entities and scenarios, LIT-RAGBench evaluates answers grounded in the provided external documents. The dataset consists of 114 human-constructed Japanese questions and an English version generated by machine translation with human curation. We use LLM-as-a-Judge for scoring and report category-wise and overall accuracy. Across API-based and open-weight models, no model exceeds 90% overall accuracy. By making strengths and weaknesses measurable within each category, LIT-RAGBench serves as a valuable metric for model selection in practical RAG deployments and for building RAG-specialized models.</abstract>
<identifier type="citekey">itai-etal-2026-lit</identifier>
<identifier type="doi">10.63317/3ep7jffajytv</identifier>
<location>
<url>https://aclanthology.org/2026.lrec-1.427/</url>
</location>
<part>
<date>2026-05</date>
<extent unit="page">
<start>5482</start>
<end>5493</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T LIT-RAGBench: Benchmarking Generator Capabilities of Large Language Models in Retrieval-Augmented Generation
%A Itai, Koki
%A Hasegawa, Shunichi
%A Yamamoto, Yuta
%A Minegishi, Gouki
%A Otsuki, Masaki
%Y Piperidis, Stelios
%Y Bel, Núria
%Y van den Heuvel, Henk
%Y Ide, Nancy
%Y Krek, Simon
%Y Toral, Antonio
%S Proceedings of the Fifteenth Language Resources and Evaluation Conference
%D 2026
%8 May
%I ELRA Language Resource Association
%C Palma de Mallorca, Spain
%F itai-etal-2026-lit
%X Retrieval-Augmented Generation (RAG) is a framework in which a Generator, such as a Large Language Model (LLM), produces answers by retrieving documents from an external collection using a Retriever. In practice, Generators must integrate evidence from long contexts, perform multi-step reasoning, interpret tables, and abstain when evidence is missing. However, existing benchmarks for Generators provide limited coverage, with none enabling simultaneous evaluation of multiple capabilities under unified conditions. To bridge the gap between existing evaluations and practical use, we introduce LIT-RAGBench (the Logic, Integration, Table, Reasoning, and Abstention RAG Generator Benchmark), which defines five categories: Integration, Reasoning, Logic, Table, and Abstention—each further divided into practical evaluation aspects. LIT-RAGBench systematically covers patterns combining multiple aspects across categories. By using fictional entities and scenarios, LIT-RAGBench evaluates answers grounded in the provided external documents. The dataset consists of 114 human-constructed Japanese questions and an English version generated by machine translation with human curation. We use LLM-as-a-Judge for scoring and report category-wise and overall accuracy. Across API-based and open-weight models, no model exceeds 90% overall accuracy. By making strengths and weaknesses measurable within each category, LIT-RAGBench serves as a valuable metric for model selection in practical RAG deployments and for building RAG-specialized models.
%R 10.63317/3ep7jffajytv
%U https://aclanthology.org/2026.lrec-1.427/
%U https://doi.org/10.63317/3ep7jffajytv
%P 5482-5493
Markdown (Informal)
[LIT-RAGBench: Benchmarking Generator Capabilities of Large Language Models in Retrieval-Augmented Generation](https://aclanthology.org/2026.lrec-1.427/) (Itai et al., LREC 2026)
ACL