<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="bin-etal-2026-knowledge">
    <titleInfo>
        <title>Knowledge Boundaries Are Not Defined by Prompts : Improving the Reachability of Knowledge through Question-Level Probing</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Wang</namePart>
        <namePart type="family">Bin</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yilu</namePart>
        <namePart type="family">Zhong</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Shuping</namePart>
        <namePart type="family">Chen</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Hui</namePart>
        <namePart type="family">LI</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>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</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">979-8-89176-459-0</identifier>
    </relatedItem>
    <abstract>Safe and reliable large language models should abstain when a question falls beyond their knowledge boundary. Existing abstention methods often use prompt-level uncertainty signals, such as semantic entropy or sampling consistency, as proxies for this boundary, treating instability under the current prompt as evidence that the question is unanswerable. We argue that this prompt-level view is overly conservative: the same underlying question can exhibit different stability under different formulations, and high uncertainty under one prompt may indicate that the answer has not been stably reached rather than that the model lacks the relevant knowledge. We propose a test-time question-level probing framework that uses existing prompt-level uncertainty signals rather than introducing a new uncertainty metric. When the original prompt is unstable, the framework probes semantically equivalent reformulations and accepts an answer only when multiple stable views provide consistent low-entropy evidence. This observable behavior serves as an operational proxy for question-level answerability under a fixed model, sampling configuration, and reformulation set. The method requires no target-model training or parameter updates and is compatible with semantic entropy and its variants. Experiments show that question-level probing recovers prompt-sensitive answers incorrectly rejected by prompt-level methods and improves the risk-coverage trade-off with controlled error rates. The code is available at https://github.com/Narwhal-Lab/Narwhal-codegen-safety-hub/tree/main/KnowledgeBoundariesAreNotDefinedByPrompts.</abstract>
    <identifier type="citekey">bin-etal-2026-knowledge</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.2109/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>47991</start>
            <end>48008</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Knowledge Boundaries Are Not Defined by Prompts : Improving the Reachability of Knowledge through Question-Level Probing
%A Bin, Wang
%A Zhong, Yilu
%A Chen, Shuping
%A LI, Hui
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F bin-etal-2026-knowledge
%X Safe and reliable large language models should abstain when a question falls beyond their knowledge boundary. Existing abstention methods often use prompt-level uncertainty signals, such as semantic entropy or sampling consistency, as proxies for this boundary, treating instability under the current prompt as evidence that the question is unanswerable. We argue that this prompt-level view is overly conservative: the same underlying question can exhibit different stability under different formulations, and high uncertainty under one prompt may indicate that the answer has not been stably reached rather than that the model lacks the relevant knowledge. We propose a test-time question-level probing framework that uses existing prompt-level uncertainty signals rather than introducing a new uncertainty metric. When the original prompt is unstable, the framework probes semantically equivalent reformulations and accepts an answer only when multiple stable views provide consistent low-entropy evidence. This observable behavior serves as an operational proxy for question-level answerability under a fixed model, sampling configuration, and reformulation set. The method requires no target-model training or parameter updates and is compatible with semantic entropy and its variants. Experiments show that question-level probing recovers prompt-sensitive answers incorrectly rejected by prompt-level methods and improves the risk-coverage trade-off with controlled error rates. The code is available at https://github.com/Narwhal-Lab/Narwhal-codegen-safety-hub/tree/main/KnowledgeBoundariesAreNotDefinedByPrompts.
%U https://aclanthology.org/2026.emnlp-main.2109/
%P 47991-48008

Markdown (Informal)

[Knowledge Boundaries Are Not Defined by Prompts : Improving the Reachability of Knowledge through Question-Level Probing](https://aclanthology.org/2026.emnlp-main.2109/) (Bin et al., EMNLP 2026)

• Knowledge Boundaries Are Not Defined by Prompts : Improving the Reachability of Knowledge through Question-Level Probing (Bin et al., EMNLP 2026)

ACL

• Wang Bin, Yilu Zhong, Shuping Chen, and Hui LI. 2026. Knowledge Boundaries Are Not Defined by Prompts : Improving the Reachability of Knowledge through Question-Level Probing. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 47991–48008, Budapest, Hungary. Association for Computational Linguistics.