Daily Arxiv

This is a page that curates AI-related papers published worldwide.
All content here is summarized using Google Gemini and operated on a non-profit basis.
Copyright for each paper belongs to the authors and their institutions; please make sure to credit the source when sharing.

A Community-driven vision for a new Knowledge Resource for AI

Created by
  • Haebom

Author

Vinay K Chaudhri, Chaitan Baru, Brandon Bennett, Mehul Bhatt, Darion Cassel, Anthony G Cohn, Rina Dechter, Esra Erdem, Dave Ferrucci, Ken Forbus, Gregory Gelfond, Michael Genesereth, Andrew S. Gordon, Benjamin Grosof, Gopal Gupta, Jim Hendler, Sharat Israni, Tyler R. Josephson, Patrick Kyllonen, Yuliya Lierler, Vladimir Lifschitz, Clifton McFate, Hande K. McGinty, Leora Morgenstern, Alessandro Oltramari, Praveen Paritosh, Dan Roth, Blake Shepard, Cogan Shimzu, Denny Vrande\v{c}i c, Mark Whiting, Michael Witbrock

Outline

This paper highlights the continued relevance of the longstanding goal of creating a comprehensive and general-purpose knowledge resource, as evidenced by the Cyc project in 1984. Despite the success of existing knowledge resources such as WordNet, ConceptNet, and Wolfram|Alpha, a verifiable, general-purpose, and widely available knowledge resource remains a serious deficiency in the AI infrastructure. Large-scale language models suffer from knowledge gaps, robot planning lacks the necessary world knowledge, and factual misinformation detection still relies on a great deal of human expertise. This paper synthesizes the findings of over 50 researchers exploring these issues at a recent AAAI workshop and presents a community-driven vision for a new knowledge infrastructure. In particular, we present a promising idea of building an open engineering framework that leverages modern knowledge representation and reasoning techniques and effectively leverages knowledge modules in the context of practical applications. Such a framework should include rules and social structures that are adopted by contributors.

Takeaways, Limitations

•
Takeaways:
◦
We present a novel approach to addressing the lack of verifiable, general, and widely available knowledge resources, which has been identified as a major deficiency in AI infrastructure.
◦
We present specific measures for effective use of knowledge modules through an open engineering framework.
◦
It presents a new paradigm for knowledge resource development and evaluation by leveraging cutting-edge technologies in knowledge representation and reasoning.
◦
A community-driven approach can increase the efficiency of building and maintaining knowledge resources.
•
Limitations:
◦
There is a lack of detailed description of the specific design and implementation of the proposed open engineering framework.
◦
There may be a lack of consideration for interoperability and compatibility issues between different knowledge modules.
◦
There is a lack of empirical research on the feasibility and effectiveness of the proposed vision.
◦
There may be a lack of discussion about the maintenance and sustainability of community-led approaches.
👍
Made with Slashpage