English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

For generative AIs to be trustworthy, establishing transparent commongrounding with humans is essential. As a preparation toward human-model commongrounding, this study examines the process of model-model common grounding. Inthis context, common ground is defined as a cognitive framework shared amongagents in communication, enabling the connection of symbols exchanged betweenagents to the meanings inherent in each agent. This connection is facilitatedby a shared cognitive framework among the agents involved. In this research, wefocus on the tangram naming task (TNT) as a testbed to examine thecommon-ground-building process. Unlike previous models designed for this task,our approach employs generative AIs to visualize the internal processes of themodel. In this task, the sender constructs a metaphorical image of an abstractfigure within the model and generates a detailed description based on thisimage. The receiver interprets the generated description from the partner byconstructing another image and reconstructing the original abstract figure.Preliminary results from the study show an improvement in task performancebeyond the chance level, indicating the effect of the common cognitiveframework implemented in the models. Additionally, we observed that incrementalbackpropagations leveraging successful communication cases for a component ofthe model led to a statistically significant increase in performance. Theseresults provide valuable insights into the mechanisms of common grounding madeby generative AIs, improving human communication with the evolving intelligentmachines in our future society.

Cognitive Architecture Toward Common Ground Sharing Among Humans and Generative AIs: Trial on Model-Model Interactions in Tangram Naming Task