Carla Baudonne
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Carla Baudonne has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Agnès Helme-Guizon
University of Grenoble Alpes, Grenoble INP, CERAG
3 shared publications - Catherine Pelachaud
ISIR, CNRS, Sorbonne Université
3 shared publications - Florian Pecune
Univ. Bordeaux, CNRS, SANPSY, UMR6033 Bordeaux, France, Bordeaux, France
3 shared publications - Laurence Chaby
Université Paris Cité, Vision Action Cognition, Boulogne- Billancourt, France
3 shared publications - Lucie Galland
ISIR, Sorbonne University
3 shared publications - Nezih Younsi
ISIR - Sorbonne University
3 shared publications
Latest Publications
- 24 December 2025
- Published by MyJove Corporation
- No. 226,p. e69254
JoVE Video Dataset
The growing demand for therapeutic support increasingly exceeds the capacity of available professionals. A virtual agent capable of performing motivational interviewing (MI) offers a promising solution to assist patients in reaching their goal of behavior change between sessions with human therapists. MI is inherently a cooperative and adaptive form of communication. Therefore, developing an agent capable of adapting its conversational strategies to the context could significantly enhance the effectiveness of therapy. During MI sessions, human therapists adjust both their verbal and nonverbal behaviors based on the human patients' responses, as well as their profiles. Depending on the patient's level of motivation, the therapist will modify their approach accordingly. Thus, personalization and adaptability are essential for developing effective MI virtual agents. In this paper, we present a virtual agent capable of conducting MI sessions by dynamically adapting verbally and nonverbally to users in real time. Leveraging state-of-the-art models, this system enables MI interactions. The virtual agent is embodied using the Greta 2.0 platform. Its nonverbal behavior is generated through a diffusion model called MODIFF, which adapts to the user's facial expressions and their readiness to change. These facial expressions were learned on an MI corpus and validated through a dedicated user study. The dialogue is generated using a state-of-the-art large language model (LLM), enhanced by a dialogue manager specifically designed for MI, with a reinforcement learning approach, and validated through user testing. Furthermore, the dialogue manager is able to adapt to different user profiles. The resulting platform is open-source and facilitates the generation of real-time, multimodal MI dialogues, providing new tools for digitally mediated therapeutic interactions.
- 24 December 2025
- Published by MyJove Corporation
- No. 226,p. e69254
JoVE Video Dataset
The growing demand for therapeutic support increasingly exceeds the capacity of available professionals. A virtual agent capable of performing motivational interviewing (MI) offers a promising solution to assist patients in reaching their goal of behavior change between sessions with human therapists. MI is inherently a cooperative and adaptive form of communication. Therefore, developing an agent capable of adapting its conversational strategies to the context could significantly enhance the effectiveness of therapy. During MI sessions, human therapists adjust both their verbal and nonverbal behaviors based on the human patients' responses, as well as their profiles. Depending on the patient's level of motivation, the therapist will modify their approach accordingly. Thus, personalization and adaptability are essential for developing effective MI virtual agents. In this paper, we present a virtual agent capable of conducting MI sessions by dynamically adapting verbally and nonverbally to users in real time. Leveraging state-of-the-art models, this system enables MI interactions. The virtual agent is embodied using the Greta 2.0 platform. Its nonverbal behavior is generated through a diffusion model called MODIFF, which adapts to the user's facial expressions and their readiness to change. These facial expressions were learned on an MI corpus and validated through a dedicated user study. The dialogue is generated using a state-of-the-art large language model (LLM), enhanced by a dialogue manager specifically designed for MI, with a reinforcement learning approach, and validated through user testing. Furthermore, the dialogue manager is able to adapt to different user profiles. The resulting platform is open-source and facilitates the generation of real-time, multimodal MI dialogues, providing new tools for digitally mediated therapeutic interactions.
- 23 December 2025
- Published by MyJove Corporation in Journal of Visualized Experiments
- No. 226,p. e69254
Virtual Agent for Real-Time Motivational Interviewing by Integrating Adaptive Nonverbal Behavior and Language Models
The growing demand for therapeutic support increasingly exceeds the capacity of available professionals. A virtual agent capable of performing motivational interviewing (MI) offers a promising solution to assist patients in reaching their goal of behavior change between sessions with human therapists. MI is inherently a cooperative and adaptive form of communication. Therefore, developing an agent capable of adapting its conversational strategies to the context could significantly enhance the effectiveness of therapy. During MI sessions, human therapists adjust both their verbal and nonverbal behaviors based on the human patients' responses, as well as their profiles. Depending on the patient's level of motivation, the therapist will modify their approach accordingly. Thus, personalization and adaptability are essential for developing effective MI virtual agents. In this paper, we present a virtual agent capable of conducting MI sessions by dynamically adapting verbally and nonverbally to users in real time. Leveraging state-of-the-art models, this system enables MI interactions. The virtual agent is embodied using the Greta 2.0 platform. Its nonverbal behavior is generated through a diffusion model called MODIFF, which adapts to the user's facial expressions and their readiness to change. These facial expressions were learned on an MI corpus and validated through a dedicated user study. The dialogue is generated using a state-of-the-art large language model (LLM), enhanced by a dialogue manager specifically designed for MI, with a reinforcement learning approach, and validated through user testing. Furthermore, the dialogue manager is able to adapt to different user profiles. The resulting platform is open-source and facilitates the generation of real-time, multimodal MI dialogues, providing new tools for digitally mediated therapeutic interactions.
- 23 December 2025
- Published by MyJove Corporation in Journal of Visualized Experiments
- No. 226,p. e69254
Virtual Agent for Real-Time Motivational Interviewing by Integrating Adaptive Nonverbal Behavior and Language Models
The growing demand for therapeutic support increasingly exceeds the capacity of available professionals. A virtual agent capable of performing motivational interviewing (MI) offers a promising solution to assist patients in reaching their goal of behavior change between sessions with human therapists. MI is inherently a cooperative and adaptive form of communication. Therefore, developing an agent capable of adapting its conversational strategies to the context could significantly enhance the effectiveness of therapy. During MI sessions, human therapists adjust both their verbal and nonverbal behaviors based on the human patients' responses, as well as their profiles. Depending on the patient's level of motivation, the therapist will modify their approach accordingly. Thus, personalization and adaptability are essential for developing effective MI virtual agents. In this paper, we present a virtual agent capable of conducting MI sessions by dynamically adapting verbally and nonverbally to users in real time. Leveraging state-of-the-art models, this system enables MI interactions. The virtual agent is embodied using the Greta 2.0 platform. Its nonverbal behavior is generated through a diffusion model called MODIFF, which adapts to the user's facial expressions and their readiness to change. These facial expressions were learned on an MI corpus and validated through a dedicated user study. The dialogue is generated using a state-of-the-art large language model (LLM), enhanced by a dialogue manager specifically designed for MI, with a reinforcement learning approach, and validated through user testing. Furthermore, the dialogue manager is able to adapt to different user profiles. The resulting platform is open-source and facilitates the generation of real-time, multimodal MI dialogues, providing new tools for digitally mediated therapeutic interactions.
- Open Access
Virtual Agent for Real-Time Motivational Interviewing by Integrating Adaptive Nonverbal Behavior and Language Models
We present an open-source virtual agent platform for conducting real-time motivational interviews, combining state-of-the-art language and diffusion models to adapt to users' behavior and profile. Utilizing the Greta 2.0 platform, it supports various topics, including nutrition and sport-focused interventions, and offers a flexible, validated tool for enhancing digital therapeutic interactions. The growing demand for therapeutic support increasingly exceeds the capacity of available professionals. A virtual agent capable of performing motivational interviewing (MI) offers a promising solution to assist patients in reaching their goal of behavior change between sessions with human therapists. MI is inherently a cooperative and adaptive form of communication. Therefore, developing an agent capable of adapting its conversational strategies to the context could significantly enhance the effectiveness of therapy. During MI sessions, human therapists adjust both their verbal and nonverbal behaviors based on the human patients' responses, as well as their profiles. Depending on the patient's level of motivation, the therapist will modify their approach accordingly. Thus, personalization and adaptability are essential for developing effective MI virtual agents. In this paper, we present a virtual agent capable of conducting MI
Open AccessVirtual Agent for Real-Time Motivational Interviewing by Integrating Adaptive Nonverbal Behavior and Language Models
We present an open-source virtual agent platform for conducting real-time motivational interviews, combining state-of-the-art language and diffusion models to adapt to users' behavior and profile. Utilizing the Greta 2.0 platform, it supports various topics, including nutrition and sport-focused interventions, and offers a flexible, validated tool for enhancing digital therapeutic interactions. The growing demand for therapeutic support increasingly exceeds the capacity of available professionals. A virtual agent capable of performing motivational interviewing (MI) offers a promising solution to assist patients in reaching their goal of behavior change between sessions with human therapists. MI is inherently a cooperative and adaptive form of communication. Therefore, developing an agent capable of adapting its conversational strategies to the context could significantly enhance the effectiveness of therapy. During MI sessions, human therapists adjust both their verbal and nonverbal behaviors based on the human patients' responses, as well as their profiles. Depending on the patient's level of motivation, the therapist will modify their approach accordingly. Thus, personalization and adaptability are essential for developing effective MI virtual agents. In this paper, we present a virtual agent capable of conducting MI