FSS2023: Integration of Cognitive Architectures and Generative Models

Description of the symposium:
The purpose of this symposium is to explore possible integrations of cognitive architectures and generative models, two very different approaches for developing general embodied intelligent agents with possibly complementary strengths and weaknesses. An important goal is to understand the strengths, weaknesses, tradeoffs, and challenges among alternative approaches to integration, as well as potential directions for future research. We hope to include a broad range of researchers whose interests span cognitive architectures and generative models.

Organizing Committee:

  • John Laird, Center for Integrated Cognition/IQMRI
  • Christian Lebiere, Carnegie Mellon University
  • David Reitter, Google Deepmind
  • Paul S. Rosenbloom, University of Southern California
  • Andrea Stocco, University of Washington

Meeting Schedule:

  • The FSS is Oct 25-27 at the Westin Arlington Gateway, Arlington, Virginia, USA
  • See sections below for details on the schedule of talks, posters, and discussions
  • Papers and printable schedule (contact moderators for password)

Wed, 25 Oct.

9:00 AM Introduction (Mod: John Laird)
(30) Introduction to Structure and Themes of Symposium John Laird
(30) Integrating Large Language Models and Cognitive Architectures: Exploring Synergistic Approaches for AI Agents Oscar Javier Romero López, John Zimmerman, Aaron Steinfeld and Anthony Tomasic

(30) Cognitive Architectures for Language Agents

Theodore Sumers, Shunyu Yao,
Karthik Narasimhan, Thomas L. Griffiths
10:30 AM Break
11:00 AM Offline Integrations / Content (David Reitter)
(15) Automating Knowledge Acquisition for Content-Centric Cognitive Agents Using LLMs Sergei Nirenburg, Sanjay Sarma Oruganti Venkata, Jesse English and Marjorie McShane
(15) Generating Chunks for Cognitive Architectures Goonmeet Kaur Bajaj, Kate Pearce, Sean Kennedy, Othalia Larue, Alexander Hough, Jayde King, Christopher Myers and Srinivasan Parthasarathy
(15) Opportunities and Challenges in Applying Generative Methods to Exploring and Validating the Common Model of Cognition Catherine Sibert and Sönke Steffen
(15) Comparing LMMs for Prompt-Enhanced ACT-R and Soar Model Development: A Case Study in Cognitive Simulation Siyu Wu, Rodrigo Souza, Frank Ritter and Walter Lima
(10) Behavior Prediction using LLMs Anthony Harrison, Laura Hiatt and Greg Trafton
(30) Discussion David Reitter
12:30 PM Lunch
2:00 PM Poster Session
3:30 PM Break
4:00 PM Individual Module Extensions to Cognitive Architectures (Christian Lebiere)
(25) Exploiting Language Models as a Source of Knowledge for Cognitive Agents James Kirk, Robert Wray and John Laird
(15) Using Large Language Models in the Companion Cognitive Architecture: A Case Study and Future Prospects Constantine Nakos and Kenneth Forbus
(15) On Using Generative Models in a Cognitive Architecture for Embodied Agents Dongkyu Choi
(15) Proposal for Cognitive Architecture and Transformer Integration: Online Learning from Agent Experience John Laird, Robert Wray, Steven Jones, James Kirk and Peter Lindes
(20) Discussion Christian Lebiere
5:30 PM Break
6:00 PM Reception

Thursday, 26 Oct

9:00 AM Impact Across Multiple Traditional Cognitive Architecture Modules (Andrea Stocco)
(20) Psychologically-Valid Generative Agents: A Novel Approach to Agent-Based Modeling in Social Sciences Konstantinos Mitsopoulos, Ritwik Bose, Brodie Mather, Archna Bhatia, Kevin Gluck, Bonnie Dorr, Christian Lebiere and Peter Pirolli
(20) Bridging Generative Networks with the Common Model of Cognition Robert West, Spencer Eckler, Brendan Conway-Smith, Nico Turcas, Eilene Tomkins-Flanagan and Mary Kelly
(20) A Proposal for a Language Model Based Cognitive Architecture Kobe Knowles, Michael Witbrock, Gillian Dobbie and Vithya Yogarajan
(30) Discussion Andrea Stocco
10:30 AM Break
11:00 AM What Generative Models can do for Cognitive Architectures and Appropriate Use Cases? (Andrea Stocco)
(60) Breakout Sessions
(30) Plenary Summary of Breakout Sessions and Consolidation Andrea Stocco
12:30 PM Lunch
2:00 PM Alternative Integrations/Unifications of Generative Models and Cognitive Architecture
(Paul Rosenbloom)
(15) The Grounding Problem: An Approach to the Integration of Cognitive and Generative Models Mary Lou Maher, Dan Ventura and Brian Magerko
(15) Cognitive Architecture Toward Common Ground Sharing Among Humans and Generative AIs: Trial Modeling on Model-Model Interaction in Tangram Naming Task Junya Morita, Tatsuya Yui, Takeru Amaya, Ryuichiro Higashinaka and Yugo Takeuchi
(15) Generative Environment-Representation Instance-Based Learning: A Cognitive Model Tyler Malloy, Yinuo Du, Fei Fang and Cleotilde Gonzalez
(15) A Neuro-Mimetic Realization of the Common Model of Cognition via Hebbian Learning and Free Energy Minimization Alexander Ororbia and Mary Alexandria Kelly
(15) Bridging Cognitive Architectures and Generative Models with Vector Symbolic Algebras P. Michael Furlong and Chris Eliasmith
(15) Discussion Paul Rosenbloom
3:30 PM Break
4:00 PM Alternative Architectural Ideas and New Architectures from Generative Models/Cognitive Architectures (Paul Rosenbloom)
(60) Breakout Sessions
(30) Plenary Summary of Breakout Sessions and Consolidation

Friday, 27 October

9:00 AM What can CAs do for LLM/GAs and appropriate use cases? (John Laird)
(60) Breakout Sessions 
(30) Plenary Summary of Breakout Sessions and Consolidation
10:30 AM Break
11:00 AM Summary and discussion of next steps (Christian Lebiere)
(90) Plenary Discussion 
12:30 PM End of FSS
Posters for the AAAI Fall Symposium Integration of Cognitive Architecture and Generative Models
Authors Title
Christopher Dancy and Deja Workman On integrating Generative Models into Cognitive Architectures for Improved Computational Sociocultural Representations
Chase McDonald, Tyler Malloy, Thuy Ngoc Nguyen and Cleotilde Gonzalez Exploring the Path from Instructions to Rewards with Large Language Models in Instance-Based Learning
Anita Raja, Alisa Leshchenko, and Jihie Kim Leveraging Conflict to Bridge Cognitive Reasoning and Generative Algorithms
Indelisio Prieto and Benjamin Blakely Proposed Uses of Generative AI in a Cybersecurity-focused Soar Agent
Alessandro Oltramari Enabling High-Level Machine Reasoning with Cognitive Neuro-Symbolic Systems
Spencer Lynn, Bryan Loyall and James Niehaus Growing An Embodied Generative Cognitive Agent
Gopal Gupta Building Intelligent Systems by Combining Machine Learning and Automated Commonsense Reasoning
Kostas Hatalis, Despina Christou, Joshua Myers, Steven Jones, Keith Lambert, Adam Amos-Binks, Zohreh Dannenhauer and Dustin Dannenhauer Memory Matters: The Need To Improve Long-Term Memory In LLM-Agents
Yanfei Liu, Yuzhou Liu and Chao Shen Combining Minds and Machines: Investigating the Fusion of Cognitive Architectures and Generative Models for General Embodied Intelligence
Robert Thomson and Nathaniel Bastian Integrating Cognitive Architectures with Foundation Models: Cognitively-Guided Few-Shot Learning to Support Trusted Artificial Intelligence
Himanshu Joshi and Volkan Ustun Augmenting Cognitive Architectures with Large Language Models
Junya Morita, Tatsuya Yui, Takeru Amaya, Ryuichiro Higashinaka and Yugo Takeuchi Cognitive Architecture Toward Common Ground Sharing Among Humans and Generative AIs: Trial Modeling on Model-Model Interaction in Tangram Naming Task