Lukas Stappen doctorate (Dr.-Ing.) in multimodal deep learning was supervised by Prof. Schuller and defended with summa cum laude (engl. with the highest honours) in 2021. He received a M.Sc. in Data Science with distinction from King's College London, UK, in 2017 and joined Prof. Dobson’s group. He (co-)authored 30+ publications (1,7k+ citations, h-index=19) in the research fields of multimodal and language deep learning, with focus on human-centric systems. Further interests are denoising in-the-wild signals and machine learning in healthcare-related domains.
Recent Publications
CAR-bench: Evaluating the Consistency and Limit-Awareness of LLM Agents under Real-World Uncertainty. ACL 2026 · Outstanding Paper Award
Safeguards for Speech2Speech LLM-Assistants: A Case Study in Automotive Applications. KI 2026 · Best Poster Award
"What Are You Doing?": Effects of Intermediate Feedback from Agentic LLM In-Car Assistants During Multi-Step Processing. CHI 2026
Agent2Agent Threats in Safety-Critical LLM Assistants: A Human-Centric Taxonomy. Whitepaper
CarMem: Enhancing Long-Term Memory in LLM Voice Assistants through Category-Bounding. COLING 2025
Collaborating with LLMs Through a Voice and Graphical User Interface. MobileHCI 2025
Integrating Generative Artificial Intelligence in Intelligent Vehicle Systems. IEEE ITSC 2023
The Multimodal Sentiment Analysis in Car Reviews (MuSe-CaR) Dataset: Collection, Insights and Improvements. IEEE Transactions on Affective Computing 2023
Full scholar
Challenges & Communities
CAR-bench: CAR-bench Challenge, IJCAI-ECAI 2026. Founder along Johannes Kirmayr and Elisabeth André.
MuSe, Multimodal Sentiment Analysis Challenge, ACM Multimedia. Founder, five editions from 2020.
Some others Interspeech Computational Paralinguistics ChallengE (ComParE), The 2026 ACII Dyadic Conversations (DaiKon) Challenge
Reviewer for ICML, ACL, EMNLP, NAACL, ICASSP, IJCAI, IEEE TAC and IEEE Transactions on Multimedia.
Recognition
Outstanding Paper Award, ACL 2026. For CAR-bench.
1st place, AgentX Competition, UC Berkeley/ Google Deepmind, 2026. Against more than 1,300 teams.
Best Poster Award, KI 2026. For the Speech2Speech safeguards case study.
Hugging Face Paper of the Day, 2026. For CAR-bench.
GAIN / YouTube Inc. Best Paper Award, IEEE 21st MMSP, Kuala Lumpur, 2019. For "From Speech to Facial Activity: Towards Cross-Modal Sequence-to-Sequence Attention Networks".
EXIST Entrepreneurial Grant, Federal Ministry for Economic Affairs, 2021. Supports PhDs founding science-based, high-tech start-ups.
BMW ProMotion Fellowship, 2018. PhD funding for "Deep Learning for Intelligent Video Analysis".
King's Robotics Grant, 2017. Lab equipment, materials and KUKA robots.
German Academic Scholarship Foundation (Studienstiftung des deutschen Volkes). Awarded to the top 1% of German students. Covered King's tuition, living costs in London and summer schools.
Max Weber Programme, Elite Network of Bavaria. Mentoring, funding and study trips to Japan, Cambridge and the Vatican.
RMIT Melbourne scholarship, 2014. Full tuition for one semester of computer science, selected from more than 60 applicants.
PROMOS Study Scholarship, 2013. Six-month exchange during the BSc.