Advances in Human-Technology Interaction and Multimodal Analysis

The field of human-technology interaction is rapidly evolving, with a growing focus on creating more natural and engaging interactions between humans and machines. Recent research has explored the use of large language models (LLMs) to enhance conversational capabilities in social robots, mitigating the uncanny valley effect and improving user experience. Multimodal analysis, incorporating text, images, and audio, is also becoming increasingly important for understanding and analyzing complex phenomena, such as emotional visual scenes, news framing, and euphemism identification. Notable papers in this area have introduced new benchmarks and evaluation suites for empathetic agents, demonstrated the effectiveness of multimodal models in identifying patterns in large datasets, and shown that LLMs can emulate human normative judgments on emotional visual scenes. Overall, the field is moving towards more sophisticated and human-like interactions between humans and machines, with significant implications for areas such as social robotics, content moderation, and human-computer interaction. Noteworthy papers include Mitigating the Uncanny Valley Effect in Hyper-Realistic Robots, which investigated the use of LLMs to mitigate the uncanny valley effect in social robots, and Artificial Intelligence Can Emulate Human Normative Judgments on Emotional Visual Scenes, which demonstrated the ability of multimodal models to emulate human emotional ratings.

Sources

Mitigating the Uncanny Valley Effect in Hyper-Realistic Robots: A Student-Centered Study on LLM-Driven Conversations

EmpathyAgent: Can Embodied Agents Conduct Empathetic Actions?

Leveraging OpenFlamingo for Multimodal Embedding Analysis of C2C Car Parts Data

Meme Similarity and Emotion Detection using Multimodal Analysis

Behaviorally Correct Learning from Informants

Clarifying Misconceptions in COVID-19 Vaccine Sentiment and Stance Analysis and Their Implications for Vaccine Hesitancy Mitigation: A Systematic Review

Unsupervised Acquisition of Discrete Grammatical Categories

Artificial Intelligence Can Emulate Human Normative Judgments on Emotional Visual Scenes

Exploring the Effect of Robotic Embodiment and Empathetic Tone of LLMs on Empathy Elicitation

Anti Robot Speciesism

Multi-Modal Framing Analysis of News

Keyword-Oriented Multimodal Modeling for Euphemism Identification

Comparative Analysis of Image, Video, and Audio Classifiers for Automated News Video Segmentation

Using AI to Summarize US Presidential Campaign TV Advertisement Videos, 1952-2012

Built with on top of