United States

The use of dialogue agents with a strategy that focuses beyond a static depth level and gradually deepens the conversation over time can nurture relationships with users. Thus, this study aimed to develop a dialogue agent capable of gradually deepening conversations over time and consequently examine its effectiveness. First, we constructed a topic depth estimation model capable of estimating the depth of any given topic. Subsequently, we developed a dialogue agent designed to gradually deepen conversation topics as interactions progress, tailored for operations on users&#39; smartphones. To evaluate the effectiveness of the developed agent, we conducted a 10-day dialogue experiment with 26 participants. The results indicated that the proposed agent elicited deep self-disclosure from users without significantly increasing psychological reluctance compared to the control condition. Moreover, the participants perceived a deeper relationship with the agent and were less likely to consider it as a non-human entity. These findings suggest that the proposed dialogue agent was capable of effectively engaging in deeper conversations over time, thereby enhancing human-agent interactions.

IEEE RO-MAN 2024

Deepening Conversations Over Time: A Chatbot with a Topic Depth Estimation Model for Gradually Engaging in Deeper Chats

linguistic communication and dialogue

creating human-robot relationships

long-term experience and longitudinal hri studies

**Welcome to EMNLP 2023!**

On behalf of the EMNLP 2023 Organizing Committee, I extend a warm and heartfelt welcome to all of you to the 2023 Conference on Empirical Methods in Natural Language Processing (EMNLP). It is with immense pleasure and excitement that we gather here in Singapore, a vibrant hub of innovation and technological advancement.

The conference program is packed with insightful presentations, thought-provoking workshops, and engaging networking opportunities. In addition to the technical sessions, we have also planned several social events that will provide you with the opportunity to connect with your colleagues.

The conference is held in person at the Resorts World Convention Centre in Singapore, and available on-line with the help of Underline.

For those in Singapore, I hope that you will find time to explore the exciting Sentosa Island, Gardens by the Bay, Singapore Botanic Gardens and the many other attractions unique to Singapore.

Of course, we are grateful to our sponsors and partners for their generous support of EMNLP 2023, whose contributions make it possible for us to host this world-class event.

Yuji Matsumoto (RIKEN AIP) 
EMNLP 2023 General Chair

To access the **EMNLP 2023** event page on Underline, you need to register for the Conference. 
Please follow **[this link](https://2023.emnlp.org/registration/)** for more details.

EMNLP 2023

EMNLP 2023 took place in Singapore from Dec 6th to Dec 10th, 2023.

The use of dialogue agents with a strategy that focuses beyond a static depth level and gradually deepens the conversation over time can nurture relationships with users. Thus, this study aimed to develop a dialogue agent capable of gradually deepening conversations over time and consequently examine its effectiveness. First, we constructed a topic depth estimation model capable of estimating the depth of any given topic. Subsequently, we developed a dialogue agent designed to gradually deepen conversation topics as interactions progress, tailored for operations on users' smartphones. To evaluate the effectiveness of the developed agent, we conducted a 10-day dialogue experiment with 26 participants. The results indicated that the proposed agent elicited deep self-disclosure from users without significantly increasing psychological reluctance compared to the control condition. Moreover, the participants perceived a deeper relationship with the agent and were less likely to consider it as a non-human entity. These findings suggest that the proposed dialogue agent was capable of effectively engaging in deeper conversations over time, thereby enhancing human-agent interactions.

technical paper

[![](https://assets.underline.io/markdown_image/1/image/f25361a85b7c83a7476c0b3c1c0bb1b2.png)](https://drive.google.com/file/d/18jcAcciCBQHj6R07Dbpm2HCOBEIwAOCU/view?usp=sharing)


Dear Colleagues and Friends,

The 33rd IEEE International Conference on Robot and Human Interactive Communication, IEEE RO-MAN 2024, was held at Pasadena Convention Center in Pasadena, USA.

The theme of the IEEE RO-MAN 2024 was **“Embracing Human-Centered HRI”**.

With the continuous technological effort of understanding human needs, creating human-desired robotic systems, and developing real human user experiences, we are marching together towards building a promising and excellent human quality of life, all for our wellness and our next generation.

The conference provided a forum for researchers and practitioners to discuss a variety of topics related to the broad theme of Robot and Human Interactive Communication involving theories, methodologies, technologies, empirical and experimental studies, and generated more insights for the future of Robot and Human Interactive Communication!

We hope you continue to enjoy the post-conference content on Underline!

You need to log in with the email address you registered with. Access credentials have been sent to your email. 

Please be sure to check your spam and other email folders if you do not see an email confirmation right away.

Please log in to explore this event.

It looks like you are not registered for this event. 

A registration is required to access this event page. 



Registration Is Required

The 33rd IEEE International Conference on Robot and Human Interactive Communication was held in Pasadena, California, USA. The theme of the IEEE RO-MAN 2024 was “Embracing Human-Centered HRI”.

This work introduces a robotics platform which comprehensively integrates multi-step action planning, natural language understanding, and memory to interactively perform service tasks in accordance with variable needs and intentions of users. The proposed architecture is built around an AI agent, derived from GPT-4, which has been embedded in an embodied system. Our approach utilizes semantic matching, plan validation, and state messages to ground the agent in the physical world, enabling a seamless merger between communication and behavior. We demonstrated the advantages of this system with an HRI study comparing mobile robots with and without conversational AI capabilities in a free-form tour-guide scenario. The increased adaptability of the system is measured along five dimensions: flexible tasking planning, interactive exploration of information, emotional friendliness, personalization, and increased overall user satisfaction.

Dobby: A Conversational Service Robot Driven by GPT-4

For robots to effectively collaborate with human partners, they need to be able to understand what instructions and/or statements mean in the context of their environment. To ground objects in a dynamic world, robots need an understanding of how both spatial and temporal relationships evolve as a function of time in the environment. However, it is computationally intensive to classify, track, and predict the motion of all objects. Approaches based on Language-Guided Temporally Adaptive Perception (LGTAP) utilize information embedded in the instruction to selectively classify objects to construct minimal but sufficiently detailed models of the environment for symbol grounding. Such methods, however, fail when the instruction refers to the future state of the environment as it lacks any notion of whether to and for how long a future prediction is necessary to ground the instruction. This prompts a reformulation of LGTAP that can selectively utilize information from past observations to accurately predict the future state of objects. This paper describes a novel approach for LGTAP for instructions that may refer to the past, present, and/or future state of the environment by closing the loop around symbol grounding and adaptive perception. A detailed analysis of a grounding problem that refers to the future state of the environment, a corpus-based analysis of performance, and a physical demonstration of natural language understanding is presented along with a description of this novel architecture.

Grounding Language Instructions That Refer to the Past, Present, And/or Future State of Objects for Human-Robot Interaction

This paper proposes an energy consumption estimation in the trajectory planning task of an industrial robot (IR) arm manipulator. Human-robot collaborative scheduling has become integral to modern manufacturing. Effective scheduling of human-robot cooperation is crucial for enhancing production efficiency. However, energy-efficient trajectory planning in human-robot interaction remains an unexplored area. To address this, a 6-DoF (Degree of Freedom) robot arm, available in the laboratory, was utilized to assess the energy consumption of this specific robot type. However, his approach can be applied to any robot, whether it's a human-robot collaboration, a mobile robot, or any other articulated robot arm manipulator. For this goal the electrical and mechanical losses of a 6-DoF robot arm manipulator from the company Universal Robots is modeled and its power losses cost function (PLCF) is derived correspondingly to calculate the electromechanical energy losses. The objective of this paper is to estimate the energy losses in a trajectory planning task in cartesian space for a tool center point (TCP) of the robot, starting from an initial point and resting at a final point. This goal is achieved by examining and analyzing the kinematics and dynamics followed by measuring the velocity and acceleration related terms of each motor at each joint of the UR3e robot arm in a laboratory setup. The results estimate the changes of energy losses in the overall trajectory following of the robot manipulator and presents the results of the estimated and measured reduction terms.

On Experimental Energy Consumption Estimation of a 6-DoF Industrial UR3e Robot Arm Manipulator in Trajectory Planning

Functional electrical stimulation is a promising technique for restoring arm function to those with paralysis from a high spinal cord injury. While simple controllers are easy to implement, model-based controllers are likely better equipped to leverage the arm's kinematic and dynamic complexity, particularly for the high variations associated with functional arm movement. One modelling technique for a model-based controller is Gaussian Process Regression. Previous simulation work has shown promise leveraging whole-arm error data to identify the arm's various subsystems, but used perfect simulated data. We asked caregivers to correct a robotic arm's movement as simulated muscles generated torque. The simulated muscles were controlled as if they were electrically stimulated human arm muscles. This study demonstrates non-expert caregivers' ability to collect this error data via whole-arm corrections, and provides insight into their ability to improve arm subsystem models made with Gaussian Process Regression. Despite significant error in caregivers' ability to provide force corrections to hold the robot in a static configuration, these corrections were leveraged to significantly improve muscle models; the muscles that improved the most were the ones primarily used to move the physiologically actuated robot.

Non-Expert Caregivers to Improve the Identification of a Physiologically Actuated Robot

In this work, we present a learning from demonstration solution for automating a surgical blunt dissection task. In addition to learning motion trajectories, our goal is to learn variable impedance behaviors that enable the robot to interact safely and compliantly during the task. To that end, we propose a teaching interface using bilateral teleoperation, which allows the natural transfer of human motions and impedance behaviors skills to robots. The demonstrated profiles are captured with Dynamic Movement Primitives and Gaussian Mixture Models, which subsequently provide the robot a reference motion plan, and a stiffness adaptation policy, during physical interaction. Experimental validation on real robot hardware shows the effectiveness of the proposed approach in terms of ensuring successful task execution, as well as safety compared to stiff high-gain control.

Learning from Demonstration of Robot Motions and Stiffness Behaviors for Surgical Blunt Dissection

Social robots and virtual agents can provide a low-threshold social interface for the independent self-health monitoring of older adults, thereby supporting their self-care and empowering them to take charge of their own health. In this work, we present a study on the interface preference of 35 older adults for a health monitoring system which they will be evaluating for eight weeks in their own homes. We found that participants who prefer the social robot have a higher affinity for technology interaction (ATI) score and are more likely to use assistance systems in their daily lives, while participants with a slightly lower ATI score prefer the virtual agent. Participants cited more personality and an interest in robots as reasons for the robot preference, and space and flexibility as reasons to prefer the virtual agent. These results underscore the importance of a personalized introduction of social technologies for health monitoring into the daily lives of older adults.

To Be or Not to (Physically) Be? A Study on Preferences in Embodied Socially Interactive Agents for Health Monitoring of Older Adults

Past research with participants in the United States showed that, in competitive group tasks, they have more positive attitudes and behaviors toward robots on their team over humans in another team. Here we present a study in which two Japanese students and two robots, placed in a randomly assigned group, compete with another student-and-robot team in a digital game. We explored participants moral behavior towards the robots, measured through their assignment of loud noise blasts to human and robot participants, and their perceptions of and attitudes towards the robots. We then compared this data to that which was collected within the United States. Results indicated that participants in Japan favored their ingroup humans and robots over outgroup agents and differentiated ingroup members more than outgroup members, as within the US. Japanese participants also anthropomorphized robots more than US participants and treated them more positively than US participants.

Ties That Bind: Group Effects in Human-Robot Team Interaction in Japan and the United States

Though some matters of consensus have begun to crystallize, scholars in human-robot interaction have thus far reasoned about artificial identity under many different definitions. Many of these seemingly disparate perspectives may, however, be unified into one coherent model through a synthesis of contemporary scientific and Buddhist philosophy of identity. Under this model, artificial and human identity are modeled equivalently under an assumption that there is no &quot;unchanging essence&quot; which constitutes an agent's identity, but rather that identity may be defined as the sum of overlapping aggregates subject to change through time. The model reckons with the idea that much of what is conceived of as identity may be arbitrarily ascribed, artificial boundaries, but that these boundaries often constitute substantial social and psychological realities. This thinking is congruent with contemporary philosophical perspectives across disciplines from biology to cognitive science. The model may serve as a useful tool for reasoning about identity in complex, dynamic situations and provide a firm foundation for work which utilizes artificial identity. The model may even offer one or two possible answers to the question: Who is a robot?

Who is a robot? A fundamental model of artificial identity

This study investigates how the level of physical anthropomorphism in conversational agents affects their persuasiveness. The experiment compared robots with low and high levels of anthropomorphism, akin to human appearance. Results indicated that while sophisticated physical embodiment did not universally enhance persuasiveness, it did have an interaction effect based on issue involvement. For low involvement issues, persuasiveness increased with higher anthropomorphism. In contrast, for high involvement issues, persuasiveness remained constant across different levels of anthropomorphism but dropped when robots closely resembled humans, possibly due to trust in high-tech robots. The findings highlight the importance of digital agents persuasive power and ethics in employing and designing robots.

The Impact of Physical Anthropomorphism in Social Robots on User Compliance: The Moderating Effect of Issue Involvement

The seamless integration and acceptance of humanoid service robots in society requires a general understanding of those robots. An overall transparent design can be a means to establish this. To guide the implementation of effective transparency mechanisms, we propose the novel Transparency Design Model, which builds on the concept of Interaction Transparency from our previous work. This model encompasses a wide range of transparency aspects aimed to improve the overall transparency of humanoid service robot designs. To demonstrate the applicability of our model and gather initial insights on its impact on robot perception, we conducted a user study involving 50 participants. The study compared a neutral and transparent robot behavior within a complex sales service scenario. Our findings indicate that the transparent design significantly improve the robots likeability, anthropomorphism, and benevolence. Additionally, we observed positive trends in attractiveness, stimulation, and novelty as measured by the User Experience Questionnaire. These measurable effects highlight the important role of Interaction Transparency. The results also imply that a holistic system analysis in multiple complex and application-dependent interactions is worthwhile as the service content influences the perception and effects of the individual design steps. Further, there is a need to develop a new scale for measuring Interaction Transparency. Lastly, future research could benefit from employing multi-modal assessment tools.

Premium content

Downloads

Next from IEEE RO-MAN 2024

Dobby: A Conversational Service Robot Driven by GPT-4

Similar lecture

Mind the Gap Between Conversations for Improved Long-Term Dialogue Generation