F e e d A I d e FeedAIde : Guiding App Users to Submit Rich Feedback Reports by Asking Context-Aware Follow-Up Questions Conference: IEEE/ACM 9th International Conference on Mobile Software Engineering and Systems; April 12, 2026; Rio de Janeiro, Brazil IEEE/ACM 9th International Conference on Mobile Software Engineering and Systems (MOBILESoft ’26), April 12, 2026, Rio de Janeiro, Brazil DOI: 10.1145/3795077.3795120 ISBN: 979-8-4007-2486-2/2026/04 CCS: Software and its engineering Requirements analysis CCS: Software and its engineering Software evolution
Ali Ebrahimi Pourasad, Meyssam Saghiri, Walid Maalej · 2026
Read F e e d A I d e FeedAIde : Guiding App Users to Submit Rich Feedback Reports by Asking Context-Aware Follow-Up Questions Conference: IEEE/ACM 9th International Conference on Mobile Software Engineering and Systems; April 12, 2026; Rio de Janeiro, Brazil IEEE/ACM 9th International Conference on Mobile Software Engineering and Systems (MOBILESoft ’26), April 12, 2026, Rio de Janeiro, Brazil DOI: 10.1145/3795077.3795120 ISBN: 979-8-4007-2486-2/2026/04 CCS: Software and its engineering Requirements analysis CCS: Software and its engineering Software evolution by Ali Ebrahimi Pourasad, Meyssam Saghiri, Walid Maalej (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-04244.txt.
Introduction
Pourasad and colleagues present FeedAIde, guiding app users to submit rich feedback reports by asking context-aware follow-up questions. User feedback is essential for mobile app success, yet user reports and developer needs often differ. They built an iOS framework of FeedAIde and tested it with users of a gym's app. Against the app's basic feedback form, participants rated FeedAIde easier and more helpful for reporting feedback.
Catalog record
| CREATOR | Ali Ebrahimi Pourasad, Meyssam Saghiri, Walid Maalej |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 51232 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.04244
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
F e e d A I d e FeedAIde : Guiding App Users to Submit Rich Feedback Reports by Asking Context-Aware Follow-Up Questions Conference: IEEE/ACM 9th International Conference on Mobile Software Engineering and Systems; April 12, 2026; Rio de Janeiro, Brazil IEEE/ACM 9th International Conference on Mobile Software Engineering and Systems (MOBILESoft ’26), April 12, 2026, Rio de Janeiro, Brazil DOI: 10.1145/3795077.3795120 ISBN: 979-8-4007-2486-2/2026/04 CCS: Software and its engineering Requirements analysis CCS: Software and its engineering Software evolution Ali Ebrahimi Pourasad, Meyssam Saghiri, Walid Maalej Abstract . User feedback is essential for the success of mobile apps, yet what users report and what developers need often diverge. Research shows that users often submit vague feedback and omit essential contextual details. This leads to incomplete reports and time-consuming clarification discussions. To overcome this challenge, we propose F e e d A I d e FeedAIde , a context-aware, interactive feedback approach that supports users during the reporting process by leveraging the reasoning capabilities of Multimodal Large Language Models. F e e d A I d e FeedAIde captures contextual information, such as the screenshot where the issue emerges, and uses it for adaptive follow-up questions to collaboratively refine with the user a rich feedback report that contains information relevant to developers. We implemented an iOS framework of F e e d A I d e FeedAIde and evaluated it on a gym’s app with its users. Compared to the app’s simple feedback form, participants rated F e e d A I d e FeedAIde as easier and more helpful for reporting feedback. An assessment by two industry experts of the resulting 54 reports showed that F e e d A I d e FeedAIde improved the quality of both bug reports and feature requests, particularly in terms of completeness. The findings of our study demonstrate the potential of context-aware, GenAI-powered feedback reporting to enhance the experience for users and increase the information value for developers. 1. Introduction User satisfaction is a critical factor driving the success of mobile apps in an increasingly competitive market ( 32 ; 26 ; 39 ) . Consequently, user feedback is essential for developers to understand user needs and improve their apps, for example, by fixing reported bugs or adding requested features ( 28 ; 31 ) . App stores enable users to easily submit feedback as a text review and a 1-5 star rating. While user feedback has become very popular in app stores with some apps receiving thousands of reports per day ( 39 ) , prior work shows that much of the feedback submitted is rather uninformative, vague, or purely negative ( 31 ; 39 ) . Non-technical users often provide unstructured reports that lack essential contextual details ( 31 ) . This creates a gap between what users report and what developers need ( 14 ) , potentially leading to user frustration if their concerns appear not to be taken seriously ( 26 ) . Research shows that about 45% of developer responses to app store reviews ask for more details. Many users never reply to follow-up questions, and those who do often take several hours or days ( 14 ; 7 ) . Prior work has proposed enhancing App Stores to increase the overall quality of apps by better supporting developers ( 13 ) . These studies highlight that user feedback from app stores is valuable but often incomplete, leading to time-consuming clarification rounds. Advances in Generative AI (GenAI), particularly Multimodal Large Language Models (MLLMs), have revealed powerful dialogue and context reasoning capabilities that can assist the gathering of higher quality user feedback for app development ( 40 ; 50 ; 51 ) . Recent research has explored how GenAI can help developers analyze and enhance user feedback ( 6 ; 9 ; 28 ) . From the user perspective, prior approaches have also attempted to support feedback reporting ( 42 ; 45 ) , but none have utilized context data (such as device information, screenshot and interaction logs) together with GenAI capabilities to instantly guide users and simplify their reporting experience while creating more valuable, rich feedback for developers. To address this gap, we introduce FeedAIde , a novel feedback approach that enables app users to easily report feedback by answering adaptive follow-up questions in order to gather information needed by developers. When a user wants to submit feedback, FeedAIde automatically captures contextual information, such as a screenshot, and uses this context to propose several feedback options for the user to choose from. It then refines the chosen option together with the user through follow-up questions to create a detailed feedback report. Building on the generalization capabilities of MLLMs, FeedAIde is not restricted to a single type of feedback, such as bug reports or feature requests, nor to a specific app domain. Instead, it can handle diverse feedback scenarios, with follow-up questions dynamically created based on the captured context. Section 2 presents the approach and an iOS implementation, which constitutes our primary contribution. To evaluate FeedAIde , we conducted a study on a production gym app. We describe the study design in Section 3 . To assess whether FeedAIde provides an easy and helpful experience to users when submitting feedback, we performed user testing with actual app users. The participants simulated reporting four previously submitted feedback cases, once with FeedAIde and once with the app’s simple text-field form. Afterward, we surveyed participants to understand their perception on FeedAIde . The results showed that users preferred FeedAIde over the app’s standard feedback form, describing it as easier and more helpful to use, while suggesting improvements such as reduced loading times. To evaluate the value of FeedAIde for developers, two industry experts assessed 54 reports collected during the user testing. Their assessment confirms that FeedAIde produced higher-quality reports for both bug reports and feature requests, while highlighting potential areas for improvement, such as adjusting follow-up questions to uncover the root cause of an issue rather than refining a solution. Section 4 reports the evaluation results, which together with the data available in our replication package ( 41 ) , constitute our second contribution. The remainder of the paper discusses threats to validity in Section 5 , reviews related work in Section 6 , and concludes with a summary and future work on GenAI-assisted user feedback flows in Section 7 . 2. Approach We propose FeedAIde , a context-aware feedback approach that leverages MLLMs to guide users in creating rich feedback reports. The framework depicted in Figure 1 collects contextual information, such as recent user interactions and a screenshot, to understand the user’s context. When the user triggers the feedback system on their mobile device, for example through a shake gesture, FeedAIde uses this context to prompt an MLLM to generate concise, context-dependent, textual feedback predictions. The user chooses one or manually writes their feedback and the model follows up with adaptive questions to gather missing details. For example, when a