A5026993762- Social Robot Interaction and HRI
- Speech and dialogue systems
- Topic Modeling
- AI in Service Interactions
- Advanced Text Analysis Techniques
- Human-Automation Interaction and Safety
- Expert finding and Q&A systems
- Innovative Human-Technology Interaction
- Merger and Competition Analysis
- Social and Intergroup Psychology
- Working Capital and Financial Performance
- Advanced Graph Neural Networks
- Taxation and Legal Issues
- Evolutionary Psychology and Human Behavior
- Research in Social Sciences
- Multimodal Machine Learning Applications
- Natural Language Processing Techniques
KTH Royal Institute of Technology
2021-2024
Delft University of Technology
2024
An interesting application for social robots is to act as a presenter, example museum guide. In this paper, we present fully automated system architecture building adaptive presentations embodied agents. The presentation generated from knowledge graph, which also used track the grounding state of information, based on multimodal feedback user. We introduce novel way use large-scale language models (GPT-3 in our case) lexicalise arbitrary graph triples, greatly simplifying design aspect...
Intelligent agents interacting with humans through conversation (such as a robot, embodied conversational agent, or chatbot) need to receive feedback from the human make sure that its communicative acts have intended consequences. At same time, agent will also seek feedback, in order ensure her In this review article, we give an overview of past and current research on how intelligent should be able both meaningful toward humans, well understanding given by users. The covers across different...
In any system that uses structured knowledge graph (KG) data as its underlying representation, KG-to-text generation is a useful tool for turning parts of the into text can be understood by humans. Recent work has shown models make use pretraining on large amounts perform well task even with relatively small sets training specific graph-to-text task. this paper, we build concept using language to zero-shot based nothing but model's understanding triple structure from what it read. We show...
Feedback is an essential part of all communication, and agents communicating with humans must be able to both give receive feedback in order ensure mutual understanding. In this paper, we analyse multimodal given by towards a robot that presenting piece art shared environment, similar museum setting. The data analysed contains video audio recordings 28 participants, the has been richly annotated terms cues (speech, gaze, head gestures, facial expressions, body pose), as well polarity any...
The work presented here is a culmination of developments within the Swedish project COIN: Co-adaptive human-robot interactive systems, funded by Foundation for Strategic Research (SSF), which addresses unified framework co-adaptive methodologies in human–robot co-existence. We investigate co-adaptation context safe planning/control, trust, and multi-modal interactions, present novel methods that allow humans robots to adapt one another discuss directions future work.
In mid-2023, we performed an experiment in autonomous buses Stockholm, Sweden, to evaluate the role that social robots might have such settings, and their effects on passengers' feeling of safety security, given absence human drivers or clerks. To address situations may occur public transit (APT), compared embodied agent a disembodied agent. this video publication, showcase some things worked with interactions created, problematic issues had not anticipated.
People often form immediate expectations about other people, or groups of based on visual appearance and characteristics their voice speech. These stereotypes, inaccurate overgeneralized, may translate to robots that carry human-like qualities. This study aims explore if nationality-based preconceptions regarding accents can be found in people’s perception a virtual physical social robot. In an online survey with 80 subjects evaluating different first-language-influenced English...