Results, publications, deliverables
Project publications will be listed here as they are released. The consortium is committed to reaching a total of at least 25 publications in journals and peer-reviewed conferences, and all modules developed inside the project will be made available through an open-source repository by the end of the project.
Selected background publications of the consortium
- V. Sucal, M. Jullien, A. Njifenjou, F. Lefèvre, « LLM-based agentic workflow on verbal and non-verbal audiovisual perceptions and actions for proactive situated human-robot interactions » 13th International Conference on Human-Agent Interaction (HAI), 2025.
- X. Alameda-Pineda et al., « Socially pertinent robots in gerontological healthcare » International Journal of Social Robotics, 2025.
- T. Dhaussy, B. Jabaian, F. Lefèvre, « FlowAct: a proactive multimodal human-robot interaction system with continuous flow of perception and modular action sub-systems » IJCAI (under review), 2024.
- T. Parcollet et al., « LeBenchmark 2.0: a standardized, replicable and enhanced framework for self-supervised representations of French speech » Computer Speech & Language, 2024.
- A. Njifenjou, V. Sucal, B. Jabaian, F. Lefèvre, « Role-play zero-shot prompting with Large Language Models for open-domain human-machine conversation » arXiv, 2024.
- T. Deschamps-Berger, L. Lamel, L. Devillers, « Multiscale contextual learning for speech emotion recognition in emergency call center conversations » ICMI, 2023.
- T. Dhaussy, B. Jabaian, F. Lefèvre, R. Horaud, « Audio-visual speaker diarization in the framework of multi-user human-robot interaction » IEEE ICASSP, 2023.
- T. Dhaussy, B. Jabaian, F. Lefèvre, « Interaction acceptance modelling and estimation for a proactive engagement in the context of human-robot interactions » IEEE/CVF ICCV Workshops, 2023.
- F. Lefèvre, T. Dhaussy, B. Jabaian, A. Njifenjou, V. Sucal, « muDialBot, vers l’interaction humain-robot multimodale pro-active » CORIA-TALN, 2023.
- T. Cordier, L. M. Rojas-Barahona, T. Urvoy, F. Lefèvre, « Diluted near-optimal expert demonstrations for guiding dialogue stochastic policy optimisation » NeurIPS Workshop: Human in the Loop Dialogue Systems, 2020.
Deliverables
| ID | Deliverable | Due | Partners |
|---|---|---|---|
| D1.1 | Evaluation of algorithms for extracting audio and visual cues | M18 | INRIA |
| D1.2 | Evaluation of algorithms for continual audiovisual adaptation | M30 | INRIA |
| D1.3 | Methods for simulation of behavior understanding modules | M36 | INRIA, LIA |
| D2.1 | Evaluation of algorithms for automatic speech recognition and understanding | M33 | LIA |
| D2.2 | Methods for emotion detection | M33 | LISN |
| D2.3 | Evaluation of adapted algorithms in simulated environments | M36 | LISN, LIA, INRIA |
| D3.1 | Development of a vocal chatbot | M30 | LIA, LISN |
| D3.2 | A multimodal stochastic observation fusion module | M36 | LIA, INRIA |
| D3.3 | Online learning of engagement strategies | M42 | LIA, LISN, INRIA |
| D3.4 | Evaluation of adapted algorithms in simulated environments | M45 | LIA, LISN, INRIA |
| D4.1 | Report on collaborative scenarios, experiments, hardware and software specifications | M12 | AP-HP, all |
| D4.2 | Privacy and ethics guidelines for experimental validation and data collection | M12 | AP-HP, all |
| D4.3 | Report on the WoZ experiments and amended specifications | M24 | AP-HP, all |
| D4.4 | Data collected during WoZ experiments | M24 | INRIA, all |
| D4.5 | Report of each iterative evaluation | M36 | AP-HP, all |
| D4.6 | Performance of the full architecture in complex collaborative scenarios | M47 | AP-HP, all |
| D5.1 | AnandaBot website | M3 | LIA |
| D5.2 | List of scientific publications | M24, M36, M48 | LIA |
| D5.3 | Data management plan | M24 | All |
| D5.4 | Project periodic reports | M12, M24, M36, M48 | All |
| D5.5 | Project final report | M48 | All |