@inproceedings{abdi-etal-2025-hallurag,
title = "{H}allu{RAG}-{RUG} at {S}em{E}val-2025 Task 3: Using Retrieval-Augmented Generation for Hallucination Detection in Model Outputs",
author = "Abdi, Silvana and
Hassani, Mahrokh and
Kinds, Rosalien and
Strijbis, Timo and
Terpstra, Roman",
editor = "Rosenthal, Sara and
Ros{\'a}, Aiala and
Ghosh, Debanjan and
Zampieri, Marcos",
booktitle = "Proceedings of the 19th International Workshop on Semantic Evaluation (SemEval-2025)",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.semeval-1.116/",
pages = "846--851",
ISBN = "979-8-89176-273-2",
abstract = "Large Language Models (LLMs) suffer from a critical limitation: hallucinations, which refers to models generating fluent but factually incorrect text. This paper presents our approach to hallucination detection in English model outputs as part of the SemEval-2025 Task 3 (Mu-SHROOM). Our method, HalluRAG-RUG, integrates Retrieval-Augmented Generation (RAG) using Llama-3 and prediction models using token probabilities and semantic similarity. We retrieved relevant factual information using a named entity recognition (NER)-based Wikipedia search and applied abstractive summarization to refine the knowledge base. The hallucination detection pipeline then used this retrieved knowledge to identify inconsistent spans in model-generated text. This result was combined with the results of two systems which identified hallucinations based on token probabilities and low-similarity sentences. Our system placed 33rd out of 41, performing slightly below the `mark all' baseline but surpassing the `mark none' and `neural' baselines with an IoU of 0.3093 and a correlation of 0.0833."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="abdi-etal-2025-hallurag">
<titleInfo>
<title>HalluRAG-RUG at SemEval-2025 Task 3: Using Retrieval-Augmented Generation for Hallucination Detection in Model Outputs</title>
</titleInfo>
<name type="personal">
<namePart type="given">Silvana</namePart>
<namePart type="family">Abdi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mahrokh</namePart>
<namePart type="family">Hassani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rosalien</namePart>
<namePart type="family">Kinds</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Timo</namePart>
<namePart type="family">Strijbis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Roman</namePart>
<namePart type="family">Terpstra</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 19th International Workshop on Semantic Evaluation (SemEval-2025)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sara</namePart>
<namePart type="family">Rosenthal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aiala</namePart>
<namePart type="family">Rosá</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Debanjan</namePart>
<namePart type="family">Ghosh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marcos</namePart>
<namePart type="family">Zampieri</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Vienna, Austria</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-273-2</identifier>
</relatedItem>
<abstract>Large Language Models (LLMs) suffer from a critical limitation: hallucinations, which refers to models generating fluent but factually incorrect text. This paper presents our approach to hallucination detection in English model outputs as part of the SemEval-2025 Task 3 (Mu-SHROOM). Our method, HalluRAG-RUG, integrates Retrieval-Augmented Generation (RAG) using Llama-3 and prediction models using token probabilities and semantic similarity. We retrieved relevant factual information using a named entity recognition (NER)-based Wikipedia search and applied abstractive summarization to refine the knowledge base. The hallucination detection pipeline then used this retrieved knowledge to identify inconsistent spans in model-generated text. This result was combined with the results of two systems which identified hallucinations based on token probabilities and low-similarity sentences. Our system placed 33rd out of 41, performing slightly below the ‘mark all’ baseline but surpassing the ‘mark none’ and ‘neural’ baselines with an IoU of 0.3093 and a correlation of 0.0833.</abstract>
<identifier type="citekey">abdi-etal-2025-hallurag</identifier>
<location>
<url>https://aclanthology.org/2025.semeval-1.116/</url>
</location>
<part>
<date>2025-07</date>
<extent unit="page">
<start>846</start>
<end>851</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T HalluRAG-RUG at SemEval-2025 Task 3: Using Retrieval-Augmented Generation for Hallucination Detection in Model Outputs
%A Abdi, Silvana
%A Hassani, Mahrokh
%A Kinds, Rosalien
%A Strijbis, Timo
%A Terpstra, Roman
%Y Rosenthal, Sara
%Y Rosá, Aiala
%Y Ghosh, Debanjan
%Y Zampieri, Marcos
%S Proceedings of the 19th International Workshop on Semantic Evaluation (SemEval-2025)
%D 2025
%8 July
%I Association for Computational Linguistics
%C Vienna, Austria
%@ 979-8-89176-273-2
%F abdi-etal-2025-hallurag
%X Large Language Models (LLMs) suffer from a critical limitation: hallucinations, which refers to models generating fluent but factually incorrect text. This paper presents our approach to hallucination detection in English model outputs as part of the SemEval-2025 Task 3 (Mu-SHROOM). Our method, HalluRAG-RUG, integrates Retrieval-Augmented Generation (RAG) using Llama-3 and prediction models using token probabilities and semantic similarity. We retrieved relevant factual information using a named entity recognition (NER)-based Wikipedia search and applied abstractive summarization to refine the knowledge base. The hallucination detection pipeline then used this retrieved knowledge to identify inconsistent spans in model-generated text. This result was combined with the results of two systems which identified hallucinations based on token probabilities and low-similarity sentences. Our system placed 33rd out of 41, performing slightly below the ‘mark all’ baseline but surpassing the ‘mark none’ and ‘neural’ baselines with an IoU of 0.3093 and a correlation of 0.0833.
%U https://aclanthology.org/2025.semeval-1.116/
%P 846-851
Markdown (Informal)
[HalluRAG-RUG at SemEval-2025 Task 3: Using Retrieval-Augmented Generation for Hallucination Detection in Model Outputs](https://aclanthology.org/2025.semeval-1.116/) (Abdi et al., SemEval 2025)
ACL