<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="malviya-etal-2025-mst">
    <titleInfo>
        <title>MST-R: Multi-Stage Tuning for Retrieval Systems and Metric Evaluation</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Yash</namePart>
        <namePart type="family">Malviya</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Karan</namePart>
        <namePart type="family">Dhingra</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Maneesh</namePart>
        <namePart type="family">Singh</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2025-01</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 1st Regulatory NLP Workshop (RegNLP 2025)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Tuba</namePart>
            <namePart type="family">Gokhan</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Kexin</namePart>
            <namePart type="family">Wang</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Iryna</namePart>
            <namePart type="family">Gurevych</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Ted</namePart>
            <namePart type="family">Briscoe</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Abu Dhabi, UAE</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Regulatory documents are rich in nuanced terminology and specialized semantics. FRAG systems: Frozen retrieval-augmented generators utilizing pre-trained (or, frozen) components face consequent challenges with both retriever and answering performance. We present a system that adapts the retriever performance to the target domain using a multi-stage tuning (MST) strategy. Our retrieval approach, called MST-R (a) first fine-tunes encoders used in vector stores using hard negative mining, (b) then uses a hybrid retriever, combining sparse and dense retrievers using reciprocal rank fusion, and then (c) adapts the cross-attention encoder by fine-tuning only the top-k retrieved results. We benchmark the system performance on the dataset released for the RIRAG challenge (as part of the RegNLP workshop at COLING 2025). We achieve significant performance gains obtaining a top rank on the RegNLP challenge leaderboard. We also show that a trivial answering approach *games* the RePASs metric outscoring all baselines and a pre-trained Llama model. Analyzing this anomaly, we present important takeaways for future research. We also release our [code base](https://github.com/Indic-aiDias/MST-R)</abstract>
    <identifier type="citekey">malviya-etal-2025-mst</identifier>
    <location>
        <url>https://aclanthology.org/2025.regnlp-1.7/</url>
    </location>
    <part>
        <date>2025-01</date>
        <extent unit="page">
            <start>41</start>
            <end>51</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T MST-R: Multi-Stage Tuning for Retrieval Systems and Metric Evaluation
%A Malviya, Yash
%A Dhingra, Karan
%A Singh, Maneesh
%Y Gokhan, Tuba
%Y Wang, Kexin
%Y Gurevych, Iryna
%Y Briscoe, Ted
%S Proceedings of the 1st Regulatory NLP Workshop (RegNLP 2025)
%D 2025
%8 January
%I Association for Computational Linguistics
%C Abu Dhabi, UAE
%F malviya-etal-2025-mst
%X Regulatory documents are rich in nuanced terminology and specialized semantics. FRAG systems: Frozen retrieval-augmented generators utilizing pre-trained (or, frozen) components face consequent challenges with both retriever and answering performance. We present a system that adapts the retriever performance to the target domain using a multi-stage tuning (MST) strategy. Our retrieval approach, called MST-R (a) first fine-tunes encoders used in vector stores using hard negative mining, (b) then uses a hybrid retriever, combining sparse and dense retrievers using reciprocal rank fusion, and then (c) adapts the cross-attention encoder by fine-tuning only the top-k retrieved results. We benchmark the system performance on the dataset released for the RIRAG challenge (as part of the RegNLP workshop at COLING 2025). We achieve significant performance gains obtaining a top rank on the RegNLP challenge leaderboard. We also show that a trivial answering approach *games* the RePASs metric outscoring all baselines and a pre-trained Llama model. Analyzing this anomaly, we present important takeaways for future research. We also release our [code base](https://github.com/Indic-aiDias/MST-R)
%U https://aclanthology.org/2025.regnlp-1.7/
%P 41-51

Markdown (Informal)

[MST-R: Multi-Stage Tuning for Retrieval Systems and Metric Evaluation](https://aclanthology.org/2025.regnlp-1.7/) (Malviya et al., RegNLP 2025)

• MST-R: Multi-Stage Tuning for Retrieval Systems and Metric Evaluation (Malviya et al., RegNLP 2025)

ACL

• Yash Malviya, Karan Dhingra, and Maneesh Singh. 2025. MST-R: Multi-Stage Tuning for Retrieval Systems and Metric Evaluation. In Proceedings of the 1st Regulatory NLP Workshop (RegNLP 2025), pages 41–51, Abu Dhabi, UAE. Association for Computational Linguistics.