Bernhard Hilpert
Papers
2
Total Citations
22
H-Index
1
About
Bernhard Hilpert is a researcher at the intersection of affective computing, human-AI interaction, and interactive machine learning. His work explores how large language models (LLMs) can process and generate emotional signals, and how these capabilities can be leveraged to create more responsive, collaborative AI systems. In his highly cited 2023 paper "Fine-grained Affective Processing Capabilities Emerging from Large Language Models" (21 citations), Hilpert demonstrated that models like ChatGPT can perform nuanced affective computing tasks—such as emotion recognition and empathetic response generation—using zero-shot prompting alone, without any fine-tuning. This work revealed that sophisticated emotional intelligence can emerge from general-purpose language models, opening new avenues for empathetic AI. More recently, in "Closing the Teacher-Learner Loop: The Role of Affective Signals in Interactive RL" (2024), Hilpert investigates how human emotional feedback can guide reinforcement learning agents, advancing the field of Hybrid Intelligence. His research is notable for bridging computational linguistics, psychology, and interactive learning, with implications for mental health chatbots, adaptive tutoring systems, and socially aware robots. Hilpert’s contributions are shaping how machines understand and respond to human affect, making AI not just smarter, but more emotionally attuned.
Research Focus
Key Achievements
Top Papers
- 1
- 2