OntoAgentic AI: Symbolic Orchestration of Agentic and Robotic Systems

Call for Participation

Workshop at the 13th Annual Conference on Advances in Cognitive Systems (ACS 2026)

6 October 2026 ·

Informatics Forum, University of Edinburgh, Edinburgh, UK

https://www.ontoagenticai-acs.info

Presentations are welcome both in person and online.

About the Workshop

Most current agentic and neurosymbolic systems are built on generative AI, with symbolic, knowledge-based AI relegated to supporting roles. This workshop explores the alternative paradigm: a symbolic cognitive architecture holds the orchestrating role, and neural models are invoked as tools for the subtasks they perform best.

We call this stance OntoAgentic AI — agentic AI orchestrated by symbolic, knowledge-based systems, and neurosymbolic AI that is symbolic-neural rather than neural-symbolic. The name designates a field, not a system: "onto-" signals agency grounded in explicit, inspectable models of the world and of the agent's own knowledge, whatever form those models take in a given architecture. This design commitment is shared, in different forms, by work on Soar, ACT-R, Companion, ICARUS, CLARION, DIARC, MIDCA, Sigma, ARCADIA, OntoAgent, and related cognitive architectures.

A central goal of the workshop is community building: to make this shared commitment explicit, to examine where the approaches converge and diverge, and to begin coordinating a research agenda for the field. A featured application area is cognitive robotics, where the requirements of trust, verifiability, and explanation are non-negotiable.

We are especially interested in hearing from prospective adopters — practitioners and organizations that need reliable, trustworthy agentic systems in safety-critical or high-accountability domains such as robotics and healthcare.

This workshop complements the OntoAgentic AI workshop at IROS 2026 (Pittsburgh, 1 October 2026).

Topics of Interest

Topics include, but are not limited to:

• Symbolic and cognitive architectures as orchestrators of agentic systems
• Neurosymbolic integration in which the symbolic layer holds epistemic control
• LLMs as bounded tools: disambiguation, generation, perceptual interpretation, knowledge bootstrapping
• LLM-assisted knowledge acquisition and semantic validation of machine-proposed knowledge
• Cognitive robotics: ontologically grounded perception, planning, and dual-process control
• Human-robot and human-AI teaming; theory of mind and belief modeling
• Explainability grounded in the agent's actual reasoning, not post-hoc rationalization
• Formal verification of agent behavior
• Lifelong, cumulative learning versus retraining and in-context adaptation
• Comparative evaluation of symbolic and LLM orchestrators on shared tasks
• Benchmarks, shared infrastructure, and evaluation methodology for the community
• Trust, safety, and accountability in agentic and embodied systems
• Position statements and challenge problems

Submission Guidelines

We invite extended abstracts of up to 2 pages plus references, describing implemented systems, designs, experiments, positions, challenge problems, or work in progress. Ideas about an evaluation methodology suitable for OntoAgentic AI are particularly welcome.

Submissions must use the ACS 2026 format: http://www.cogsys.org/instructions/

Submissions are not anonymous and will be reviewed by the organizers and the program committee for relevance and quality.

All accepted submissions are presented as talks, with time for questions and discussion. Work at any stage of maturity is welcome, from implemented systems and experimental results to early-stage work, position statements, and challenge problems.

Authors indicate at submission whether they expect to present in person or online. This is non-binding and is used for scheduling only.

Submission portal: https://easychair.org/conferences/?conf=oaa26

Publication

Accepted abstracts will be posted on the workshop website. The workshop does not produce formal proceedings, so presenting at the workshop does not preclude publication of the work elsewhere.

An agreement is in place with the International Journal of Humanoid Robotics (World Scientific, peer-reviewed): authors of selected high-quality contributions will be invited to submit expanded full papers for a special issue covering both physical and cognitive aspects of intelligent robotic systems.

Important Dates

All deadlines are 11:59 pm GMT.

The workshop program and format will be announced closer to the date.

Committees

Organizing Committee

• Marjorie McShane — LEIA Lab, Rensselaer Polytechnic Institute — mcsham2@rpi.edu
• Sergei Nirenburg — LEIA Lab, Rensselaer Polytechnic Institute — nirens@rpi.edu
• Sanjay Oruganti — LEIA Lab, Rensselaer Polytechnic Institute — orugas2@rpi.edu

Contact

All questions about submissions should be emailed to contact@ontoagenticai-acs.info

Workshop website: https://www.ontoagenticai-acs.info

ACS 2026 conference: http://www.cogsys.org