<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="qi-etal-2026-scope-sql">
    <titleInfo>
        <title>SCOPE-SQL: Structured Semantic Credit for Reinforcement Learning of Interactive SQL Repair Agent</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Hong</namePart>
        <namePart type="family">Qi</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Junyi</namePart>
        <namePart type="family">Zhao</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Boyan</namePart>
        <namePart type="family">Li</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Juyong</namePart>
        <namePart type="family">Jiang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yuyu</namePart>
        <namePart type="family">Luo</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-10</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Findings of the Association for Computational Linguistics: EMNLP 2026</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Hung-Yi</namePart>
            <namePart type="family">Lee</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Sunipa</namePart>
            <namePart type="family">Dev</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jessy</namePart>
            <namePart type="family">Li</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Desmond</namePart>
            <namePart type="family">Elliott</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Budapest, Hungary</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-462-0</identifier>
    </relatedItem>
    <abstract>Text-to-SQL deployment is moving from one-shot query generation to interactive repair of executable but semantically incorrect SQL. In this setting, a model must revise an erroneous query under public tool feedback while preserving correct input structures. Existing reinforcement-learning signals are ill matched: outcome rewards reveal only whether the final SQL passes, whereas unconstrained process supervision may reward steps not grounded in SQL semantics. We propose SCOPE-SQL, a reinforcement-learning framework that treats SQL structure as an auditable credit space for interactive repair. It maps candidate queries and public feedback into F0–F6 modules—schema/runtime, table/join, projection, predicate, aggregation/grouping, order/limit, and nesting—and assigns action-level credit for repairing target modules, preserving stable modules, gathering evidence, and submitting when ready. Persistent low-credit regions then drive semantic refresh, generating validator-filtered frontier repair tasks. On BIRD-CRITIC-SQLite, SCOPE-SQL with semantic refresh reaches 48.8% success using a Qwen2.5-Coder-7B backbone, improving over current sota RL baselines by 4.4 and 4.2 points. Ablations attribute 4.4 and 1.8 points to process credit and semantic refresh. Trajectory analysis further shows better repair behavior: Repair+Preserve paths rise from 38.8% to 77.5%, while damage-and-recover paths fall from 43.8% to 1.2%. These results suggest that SQL-structured credit is an interpretable training signal for both policy learning and data expansion in interactive SQL repair.</abstract>
    <identifier type="citekey">qi-etal-2026-scope-sql</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-emnlp.951/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>20037</start>
            <end>20058</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T SCOPE-SQL: Structured Semantic Credit for Reinforcement Learning of Interactive SQL Repair Agent
%A Qi, Hong
%A Zhao, Junyi
%A Li, Boyan
%A Jiang, Juyong
%A Luo, Yuyu
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F qi-etal-2026-scope-sql
%X Text-to-SQL deployment is moving from one-shot query generation to interactive repair of executable but semantically incorrect SQL. In this setting, a model must revise an erroneous query under public tool feedback while preserving correct input structures. Existing reinforcement-learning signals are ill matched: outcome rewards reveal only whether the final SQL passes, whereas unconstrained process supervision may reward steps not grounded in SQL semantics. We propose SCOPE-SQL, a reinforcement-learning framework that treats SQL structure as an auditable credit space for interactive repair. It maps candidate queries and public feedback into F0–F6 modules—schema/runtime, table/join, projection, predicate, aggregation/grouping, order/limit, and nesting—and assigns action-level credit for repairing target modules, preserving stable modules, gathering evidence, and submitting when ready. Persistent low-credit regions then drive semantic refresh, generating validator-filtered frontier repair tasks. On BIRD-CRITIC-SQLite, SCOPE-SQL with semantic refresh reaches 48.8% success using a Qwen2.5-Coder-7B backbone, improving over current sota RL baselines by 4.4 and 4.2 points. Ablations attribute 4.4 and 1.8 points to process credit and semantic refresh. Trajectory analysis further shows better repair behavior: Repair+Preserve paths rise from 38.8% to 77.5%, while damage-and-recover paths fall from 43.8% to 1.2%. These results suggest that SQL-structured credit is an interpretable training signal for both policy learning and data expansion in interactive SQL repair.
%U https://aclanthology.org/2026.findings-emnlp.951/
%P 20037-20058

Markdown (Informal)

[SCOPE-SQL: Structured Semantic Credit for Reinforcement Learning of Interactive SQL Repair Agent](https://aclanthology.org/2026.findings-emnlp.951/) (Qi et al., Findings 2026)

• SCOPE-SQL: Structured Semantic Credit for Reinforcement Learning of Interactive SQL Repair Agent (Qi et al., Findings 2026)

ACL

• Hong Qi, Junyi Zhao, Boyan Li, Juyong Jiang, and Yuyu Luo. 2026. SCOPE-SQL: Structured Semantic Credit for Reinforcement Learning of Interactive SQL Repair Agent. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 20037–20058, Budapest, Hungary. Association for Computational Linguistics.