People communicate how things should move by combining words with demonstrations: "open it like this." We present an interaction technique that brings this expressive resource to conversational 3D authoring. Building on "Put-That-There," our system combines speech, pointing, and spatiotemporal demonstrations to specify editable behavior. A hand movement supplies evidence for a mechanism’s axis, pivot, range, and pace; an animated preview makes the interpretation inspectable. Users refine behavior through further words or demonstrations, and the system can request a demonstration to clarify intent. In a twelve-participant study comparing three input configurations, our system achieved 89% participant-declared completion versus 47% with speech alone, had the highest observed match rates on six categorical accuracy measures and tied on two, and was preferred by nine participants. This work makes demonstration part of an ongoing authoring conversation: behavior can be shown, inspected, and revised.
@article{elalaoui2026justlikethis,title={“Just Like This”: Manner Deixis at the Graphics Interface},author={El Alaoui, Hamza and Bigham, Jeffrey P. and Rekimoto, Jun},journal={arXiv preprint arXiv:2610.09099},year={2026},eprint={2610.09099},archiveprefix={arXiv},primaryclass={cs.HC},doi={10.48550/arXiv.2610.09099},}
CHI
“I followed what felt right, not what I was told”: Autonomy, Coaching, and Recognizing Bias Through AI-Mediated Dialogue
Atieh
Taheri, Hamza
El Alaoui, Patrick
Carrington, and Jeffrey P.
Bigham
In Proceedings of the 44th ACM Conference on Human Factors in Computing Systems, ACM, Barcelona, Spain, 2026
Ableist microaggressions remain pervasive in everyday interactions, yet interventions to help people recognize them are limited. We present an experiment testing how AI-mediated dialogue influences recognition of ableism. 160 participants completed a pre-test, intervention, and a post-test across four conditions: AI nudges toward bias (Bias-Directed), inclusion (Neutral-Directed), unguided dialogue (Self-Directed), and a text-only non-dialogue (Reading). Participants rated scenarios on standardness of social experience and emotional impact; those in dialogue-based conditions also provided qualitative reflections. Quantitative results showed dialogue-based conditions produced stronger recognition than Reading, though trajectories diverged: biased nudges improved differentiation of bias from neutrality but increased overall negativity. Inclusive or no nudges remained more balanced, while Reading participants showed weaker gains and even declines. Qualitative findings revealed biased nudges were often rejected, while inclusive nudges were adopted as scaffolding. We contribute a validated vignette corpus, an AI-mediated intervention platform, and design implications highlighting trade-offs conversational systems face when integrating bias-related nudges.
@inproceedings{recognizingbias2026,title={“I followed what felt right, not what I was told”: Autonomy, Coaching, and Recognizing Bias Through AI-Mediated Dialogue},author={Taheri, Atieh and El Alaoui, Hamza and Carrington, Patrick and Bigham, Jeffrey P.},booktitle={Proceedings of the 44th ACM Conference on Human Factors in Computing Systems},pages={1--23},year={2026},isbn={9798400722783},location={Barcelona, Spain},address={New York, NY, USA},doi={10.1145/3772318.3791078},publisher={ACM},}
2025
UIST
StepWrite: Adaptive Planning for Speech-Driven Text Generation
Hamza
El Alaoui, Atieh
Taheri, Yi-Hao
Peng, and Jeffrey P.
Bigham
In Proceedings of the 38th Annual ACM Symposium on User Interface Software and Technology, ACM, Busan, Republic of Korea, 2025
People frequently use speech-to-text systems to compose short texts with voice. However, current voice-based interfaces struggle to support composing more detailed, contextually complex texts, especially in scenarios where users are on the move and cannot visually track progress. Longer-form communication, such as composing structured emails or thoughtful responses, requires persistent context tracking, structured guidance, and adaptability to evolving user intentions—capabilities that conventional dictation tools and voice assistants do not support. We introduce StepWrite, a large language model-driven voice-based interaction system that augments human writing ability by enabling structured, hands-free and eyes-free composition of longer-form texts while on the move. StepWrite decomposes the writing process into manageable subtasks and sequentially guides users with contextually-aware non-visual audio prompts. StepWrite reduces cognitive load by offloading the context-tracking and adaptive planning tasks to the models. Unlike baseline methods like standard dictation features (e.g., Microsoft Word) and conversational voice assistants (e.g., ChatGPT Advanced Voice Mode), StepWrite dynamically adapts its prompts based on the evolving context and user intent, and provides coherent guidance without compromising user autonomy. An empirical evaluation with 25 participants engaging in mobile or stationary hands-occupied activities demonstrated that StepWrite significantly reduces cognitive load, improves usability and user satisfaction compared to baseline methods. Technical evaluations further confirmed StepWrite’s capability in dynamic contextual prompt generation, accurate tone alignment, and effective fact checking. This work highlights the potential of structured, context-aware voice interactions in enhancing hands-free and eye-free communication in everyday multitasking scenarios.
@inproceedings{elalaoui2025stepwrite,author={El Alaoui, Hamza and Taheri, Atieh and Peng, Yi-Hao and Bigham, Jeffrey P.},title={StepWrite: Adaptive Planning for Speech-Driven Text Generation},year={2025},isbn={9798400720376},address={New York, NY, USA},doi={10.1145/3746059.3747610},booktitle={Proceedings of the 38th Annual ACM Symposium on User Interface Software and Technology},articleno={86},numpages={30},keywords={voice interfaces, adaptive planning, task decomposition, speech-to-text, text composition, hands-free writing, eyes-free writing, large language models},location={Busan, Republic of Korea},series={UIST '25},publisher={ACM}}
Earlier publicationsnatural language processing, energy systems, autonomous driving
2024
DarijaGenie: Learning Moroccan Arabic Through a Multimodal Chatbot
Hamza
El Alaoui, and Violetta
Cavalli-Sforza
In Proceedings of the International Conference on Arabic Language Processing, Springer, Rabat, Morocco, 2024
The rapid advancement of digital technologies has significantly influenced the landscape of language learning, introducing innovative tools and methodologies to enhance educational experiences. Among these developments, the integration of artificial intelligence and chatbot technology in language education offers a unique approach to teaching less commonly taught languages. This paper introduces “DarijaGenie,” an early version of a task-based, multimodal chatbot system designed to facilitate the learning of Moroccan Arabic. Characterized by its spoken nature, non-standard structure, and the scarcity of its resources, Darija presents distinct challenges for learners. DarijaGenie aims to address these challenges by leveraging natural language processing and artificial intelligence to simulate real-life scenarios, providing users with an interactive platform for practical language learning.
@inproceedings{el2024darijagenie,title={DarijaGenie: Learning Moroccan Arabic Through a Multimodal Chatbot},author={El Alaoui, Hamza and Cavalli-Sforza, Violetta},booktitle={Proceedings of the International Conference on Arabic Language Processing},pages={74--89},year={2024},location={Rabat, Morocco},publisher={Springer},doi={10.1007/978-3-031-79164-2_7},}
2023
Building Intelligent Chatbots: Tools, Technologies, and Approaches
Hamza
El Alaoui, Zakaria
El Aouene, and Violetta
Cavalli-Sforza
In Proceedings of the International Conference on Innovative Research in Applied Science, Engineering, & Technology, IEEE, Casablanca, Morocco, 2023
This paper aims to provide a comprehensive overview of the approaches and technologies used in building chatbots, including declarative and open-domain chatbots, and to serve as a useful resource for researchers and practitioners working in this field. We review pipeline methods, which involve the use of various techniques to process, understand, and generate natural language utterances, and end-to-end methods, which involve training a single neural network to handle all aspects of the conversation. We also examine neural generative models, which use machine learning techniques to generate responses based on input data, and retrieval-based methods, which use pre-defined responses and match them to user input. Furthermore, we review the various technologies used in building chatbots, including natural language processing (NLP) libraries, frameworks, non-cloud-based platforms, and cloud-based platforms. We provide a comparison of the strengths and limitations of these approaches and technologies and discuss the potential future directions for research in this field.
@inproceedings{el2023building,title={Building Intelligent Chatbots: Tools, Technologies, and Approaches},author={El Alaoui, Hamza and El Aouene, Zakaria and Cavalli-Sforza, Violetta},booktitle={Proceedings of the International Conference on Innovative Research in Applied Science, Engineering, & Technology},pages={1--12},year={2023},location={Casablanca, Morocco},publisher={IEEE},doi={10.1109/IRASET57153.2023.10153005},}
Deep Reinforcement Learning for Autonomous Vehicle Intersection Navigation
Badr Ben
Elallid, Hamza
El Alaoui, and Nabil
Benamar
In Proceedings of the Conference on Innovation and Intelligence for Informatics, Computing, & Technologies, IEEE, Sakheer, Bahrain , 2023
In this paper, we explore the challenges associated with navigating complex T-intersections in dense traffic scenarios for autonomous vehicles (AVs). Reinforcement learning algorithms have emerged as a promising approach to address these challenges by enabling AVs to make safe and efficient decisions in real-time. Here, we address the problem of efficiently and safely navigating T-intersections using a lower-cost, single-agent approach based on the Twin Delayed Deep Deterministic Policy Gradient (TD3) reinforcement learning algorithm. We show that our TD3-based method, when trained and tested in the CARLA simulation platform, demonstrates stable convergence and improved safety performance in various traffic densities. Our results reveal that the proposed approach enables the AV to effectively navigate T-intersections, outperforming previous methods in terms of travel delays, collision minimization, and overall cost. This study contributes to the growing body of knowledge on reinforcement learning applications in autonomous driving and highlights the potential of single-agent, cost-effective methods for addressing more complex driving scenarios and advancing reinforcement learning algorithms in the future.
@inproceedings{elallid2023deep,title={Deep Reinforcement Learning for Autonomous Vehicle Intersection Navigation},author={Elallid, Badr Ben and El Alaoui, Hamza and Benamar, Nabil},booktitle={Proceedings of the Conference on Innovation and Intelligence for Informatics, Computing, & Technologies},pages={308--313},year={2023},publisher={IEEE},location={Sakheer, Bahrain },doi={10.1109/3ICT60104.2023.10391453},}
Impact of hot arid climate on optimal placement of electric vehicle charging stations
Hamza
El Hafdaoui, Hamza
El Alaoui, and
3 more authors
Electric vehicles (EVs) are becoming more commonplace as they cut down on both fossil fuel use and pollution caused by the transportation sector. However, there are a number of major issues that have arisen as a result of the rapid expansion of electric vehicles, including an inadequate number of charging stations, uneven distribution, and excessive cost. The purpose of this study is to enable EV drivers to find charging stations within optimal distances while also taking into account economic, practical, geographical, and atmospheric considerations. This paper uses the Fez-Meknes region in Morocco as a case study to investigate potential solutions to the issues raised above. The scorching, arid climate of the region could be a deterrent to the widespread use of electric vehicles there. This article first attempts to construct a model of an EV battery on MATLAB/Simulink in order to create battery autonomy of the most widely used EV car in Morocco, taking into account weather, driving style, infrastructure, and traffic. Secondly, collected data from the region and simulation results were then employed to visualize the impact of ambient temperature on EV charging station location planning, and a genetic algorithm-based model for optimizing the placement of charging stations was developed in this research. With this method, EV charging station locations were initially generated under the influence of gas station locations, population and parking areas, and traffic, and eventually through mutation, the generated initial placements were optimized within the bounds of optimal cost, road width, power availability, and autonomy range and influence. The results are displayed to readers in a node-link network to help visually represent the impact of ambient temperatures on EV charging station location optimization and then are displayed in interactive GIS maps. Finally, conclusions and research prospects were provided.
@article{el2023impact,title={Impact of hot arid climate on optimal placement of electric vehicle charging stations},author={El Hafdaoui, Hamza and El Alaoui, Hamza and Mahidat, Salma and El Harmouzi, Zakaria and Khallaayoun, Ahmed},journal={Energies},volume={16},number={2},pages={753},year={2023},publisher={MDPI},doi={10.3390/en16020753},}
Sustainable railways for Morocco: a comprehensive energy and environmental assessment
Hamza
El Alaoui, Ahmed
Bazzi, and
3 more authors
Journal of UQU for Engineering and Architecture, Springer, 2023
This study provides a comprehensive environmental impact assessment of different train types in Morocco, including passenger-only, goods-only, and mixed-use services. Through a comparative evaluation based on climate change, respiratory effects, and acidification/eutrophication, we find that passenger-only trains exhibit the highest environmental impact across all categories due to their high energy consumption and emissions. Conversely, goods-only and mixed-use trains display markedly lower environmental impacts, emphasizing the role of service type and operational efficiency in mitigating environmental harm. Our findings underscore the need for strategic optimization in train services, particularly passenger-only operations, and open the door for comprehensive economic and environmental assessments of alternative train technologies and fuel mixtures in the context of Morocco’s decarbonization plans.
@article{el2023sustainable,title={Sustainable railways for Morocco: a comprehensive energy and environmental assessment},author={El Alaoui, Hamza and Bazzi, Ahmed and El Hafdaoui, Hamza and Khallaayoun, Ahmed and Lghoul, Rachid},journal={Journal of UQU for Engineering and Architecture},volume={14},number={4},pages={271--283},year={2023},publisher={Springer},doi={10.1007/s43995-023-00034-0},}