Student thesis: Multimodal inference of user satisfaction in dialogue systems

Life is always about becoming... In life it is about going on a journey to become the best version of our future self. As we discover new things, we face challenges, master them and grow beyond us.

Apply to Mercedes-Benz and find the area where you can develop your talents individually. You will be supported by visionary colleagues who share your pioneering spirit. Joining us means becoming part of a global team whose goal is to build the most desirable cars in the world. Together for excellence.

Number: MER00043MZ Mercedes-Benz Group AG is one of the most successful automotive companies in the world. With Mercedes-Benz AG and Mercedes-Benz Cars, Mercedes-Benz Vans and Mercedes-Benz Mobility AG, the vehicle manufacturer is one of the largest and most successful providers of premium cars. Being part of Mercedes-Benz means finding the task area where you can develop your talents individually. It means in a global car company that has the goal of building the most desirable cars in the world to give its best. You will be supported by visionary colleagues who share their pioneering spirit. Together for excellence. We are Innovative Experience In our cross-functional team „Holistic Customer Centered Experiences“ we design the vision of our UX Vision for the Mercedes-Benz user experience of the future. With this vision we develop a new, holistic customer experience in our vehicles and beyond. Our customers are at the heart of the Mercedes-Benz Experience. These challenges come to you: Modern task-oriented dialogue agents rely heavily on explicit user feedback or metrics to complete tasks in order to evaluate the success of an interaction. However, explicit feedback is extremely rare in real environments, such as in-vehicle assistants in the automotive sector. This work addresses the inference of implicit, multimodal feedback as a central problem of machine learning. The focus is on the algorithmic tracking of the latent user state under partial observability. Due to the early prediction of dissatisfaction in the temporal interaction process, the model allows a closed AI system to proactively adapt its dialogue policy before the interaction fails. The primary scientific contribution will be the training of the model and the processing of a pipeline for its evaluation from machine learning sight. Your duties Development of latent models: building and training of recurrent Belief-State models over the temporal course of interaction as well as their benchmarking against purely text-based LLMs and multimodal late fusion classifiers Design of rigorous ML evaluations: implementation of evaluation pipelines with strong focus on early inference Insulation of causal signals: training of negative control variables (e.g. current road context or basic mood) in order to demonstrate mathematically that the model actually insulates the causal influence of the assistant on user satisfaction and does not merely represent environmental noise The final topics will be dealt with in consultation with the university, you and us The activity can begin from August 2026.

enrolled in the course of study: (media) informatics, artificial intelligence, machine learning, data science, engineering psychology or a comparable degree course with a corresponding focus Knowledge of Artificial Intelligence and Machine Learning First experiences in the analysis of multimodal data sets Great interest in technological trends and their potentials fun of independent work and challenges, recognizable by high motivation, willingness to implement, commitment, team capacity and active communication and analytical thinking and strategic work Secure knowledge of German and English in word and writing Safe handling of MS Office and SW development tools such as Atlassian Confluence, Jira, etc. Ideally a driving licence class B Additional information: We are looking forward to your online application with CV, lettering, certificates, current enrollment certificate with an indication of the semester and proof of the regular study period. Please do not forget to mark your documents as ''relevant for this application' in the online form and to observe the maximum file size of 5 MB. Further information on the setting criteria can be found here. Disabled and equalized applicants are welcome! The severely disabled representative (sbv-sindelfingen@mercedes-benz.com) is happy to support you in the application process. HR Services will be happy to help you with questions about the application process. You can reach us by email via myhrservice@mercedes-benz.com or by phone at 0711/17-99000 (Mo-Fr 10-12am & 13-15am).

Food supplements Employee handy possible Employee discounts possible Employee participation possible Staff Events Coaching Flexible working time possible Hybrid work possible Health measures Employment Mobility offers Parking space Business doctor Good connection Accessibility Child care Kantine, Café