@inproceedings{cho-etal-2026-act,
title = "Why Act Without Asking? Evaluating Mobile {GUI} Agents under Ambiguous Instructions",
author = "Cho, Yeongjae and
Park, Jungwuk and
Choi, Tae-Min and
Park, Chiyoun and
Kim, Mincheol and
Kim, Kookhoi",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.1236/",
pages = "28189--28210",
ISBN = "979-8-89176-459-0",
abstract = "Mobile GUI agents often act confidently under ambiguous or incomplete user instructions-a common situation in real-world use. Even when explicitly offered a ``call the user'' action, open-weight agents ask for clarification in only 0.1{\%} of cases on average, instead guessing the user{'}s intent and proceeding with actions. We name this failure Act-without-Asking. Existing step-level benchmarks largely miss this failure mode because they credit any ground-truth-matching action regardless of asking. We introduce MID-Bench, a step-level ambiguity benchmark constructed by rewriting clear instructions into ambiguous ones, with a strict clarification protocol: a user simulator answering only what is asked, and a scoring rule crediting a step only when the agent both asks an appropriate question and emits the correct final action. The Hard subset applies stricter masking to the action-critical detail in the instruction, making it harder to succeed without asking. Across four open-weight GUI agents, train-free cascades in which a gate decides whether to act or ask the user for clarification substantially improve ambiguous-task performance while preserving accuracy on clear instructions; LoRA fine-tuning on MID-Bench further combines effectively with these cascades."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="cho-etal-2026-act">
<titleInfo>
<title>Why Act Without Asking? Evaluating Mobile GUI Agents under Ambiguous Instructions</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yeongjae</namePart>
<namePart type="family">Cho</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jungwuk</namePart>
<namePart type="family">Park</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tae-Min</namePart>
<namePart type="family">Choi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chiyoun</namePart>
<namePart type="family">Park</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mincheol</namePart>
<namePart type="family">Kim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kookhoi</namePart>
<namePart type="family">Kim</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>Mobile GUI agents often act confidently under ambiguous or incomplete user instructions-a common situation in real-world use. Even when explicitly offered a “call the user” action, open-weight agents ask for clarification in only 0.1% of cases on average, instead guessing the user’s intent and proceeding with actions. We name this failure Act-without-Asking. Existing step-level benchmarks largely miss this failure mode because they credit any ground-truth-matching action regardless of asking. We introduce MID-Bench, a step-level ambiguity benchmark constructed by rewriting clear instructions into ambiguous ones, with a strict clarification protocol: a user simulator answering only what is asked, and a scoring rule crediting a step only when the agent both asks an appropriate question and emits the correct final action. The Hard subset applies stricter masking to the action-critical detail in the instruction, making it harder to succeed without asking. Across four open-weight GUI agents, train-free cascades in which a gate decides whether to act or ask the user for clarification substantially improve ambiguous-task performance while preserving accuracy on clear instructions; LoRA fine-tuning on MID-Bench further combines effectively with these cascades.</abstract>
<identifier type="citekey">cho-etal-2026-act</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.1236/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>28189</start>
<end>28210</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Why Act Without Asking? Evaluating Mobile GUI Agents under Ambiguous Instructions
%A Cho, Yeongjae
%A Park, Jungwuk
%A Choi, Tae-Min
%A Park, Chiyoun
%A Kim, Mincheol
%A Kim, Kookhoi
%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 cho-etal-2026-act
%X Mobile GUI agents often act confidently under ambiguous or incomplete user instructions-a common situation in real-world use. Even when explicitly offered a “call the user” action, open-weight agents ask for clarification in only 0.1% of cases on average, instead guessing the user’s intent and proceeding with actions. We name this failure Act-without-Asking. Existing step-level benchmarks largely miss this failure mode because they credit any ground-truth-matching action regardless of asking. We introduce MID-Bench, a step-level ambiguity benchmark constructed by rewriting clear instructions into ambiguous ones, with a strict clarification protocol: a user simulator answering only what is asked, and a scoring rule crediting a step only when the agent both asks an appropriate question and emits the correct final action. The Hard subset applies stricter masking to the action-critical detail in the instruction, making it harder to succeed without asking. Across four open-weight GUI agents, train-free cascades in which a gate decides whether to act or ask the user for clarification substantially improve ambiguous-task performance while preserving accuracy on clear instructions; LoRA fine-tuning on MID-Bench further combines effectively with these cascades.
%U https://aclanthology.org/2026.emnlp-main.1236/
%P 28189-28210
Markdown (Informal)
[Why Act Without Asking? Evaluating Mobile GUI Agents under Ambiguous Instructions](https://aclanthology.org/2026.emnlp-main.1236/) (Cho et al., EMNLP 2026)
ACL