About
Recent advances in robotics have significantly improved perception and policy learning through data-driven approaches, including learning from demonstrations, world models, and vision–language–action (VLA) models. These methods enable robots to acquire complex behaviors from large-scale data and have expanded the scope of tasks that robots can perform in unstructured environments. However, deploying robotic systems reliably in real-world settings such as homes, manufacturing facilities, and public infrastructure remains challenging. Robots must reason about task structure, physical constraints, and safety requirements, and adapt to changing environments and incomplete information in real-time.

Embodied Neuro-Symbolic (NeSy) AI explores how neural learning can be combined with structured knowledge representations and symbolic reasoning to support reliable decision-making in physical AI systems. In such systems, robots must be able to (i) ground perceptual observations into structured representations capturing objects, relations, and task-relevant constraints, (ii) reason over explicit knowledge about skills, tasks, environments, and operational/safety requirements to generate feasible and verifiable action plans, and (iii) adapt behavior through learning while maintaining interpretable and constraint-aware decision processes.

This workshop aims to foster discussion on how neuro-symbolic mechanisms can enable more reliable, adaptable, and trustworthy intelligence for robotic systems.

Intended audience. This workshop targets engineers, researchers, and practitioners interested in deploying reliable, task- and constraint-aware robots in complex, real-world environments. It aims to foster interdisciplinary collaboration across robot learning, cognitive robotics, planning and control, embodied AI, and neuro-symbolic AI, as well as related fields such as computer vision, natural language processing, and multimodal machine learning.

Invited Speakers

Schedule

Time
08:25		Organizers Introductory Remarks
08:30		Keynote 1: Alessandra Sciutti Title TBD Abstract Abstract TBD
08:50		Keynote 2: Sebastian Scherer Title TBD Abstract Abstract TBD
09:10		Keynote 3: Jiayuan Mao Title TBD Abstract Abstract TBD
09:30		Oral Talks
09:50		Keynote 4: Yezhou (YZ) Yang Title TBD Abstract Abstract TBD
10:10		Keynote 5: Joyce Chai Title TBD Abstract Abstract TBD
10:30		Coffee Break (+ Poster Session)
11:00		Keynote 6: Paulo Shankarian Title TBD Abstract Abstract TBD
11:20		Debate (all speakers) - What semantics must the data capture for efficient robot learning, and how important is direct human input in generating and curating that data?
12:25		Organizers Best Paper Award + Closing Remarks

Call for Papers

Targeted Topics

We invite paper submissions on various topics related to Embodied Neuro-Symbolic AI for robotics, including (but not limited to):

• Neuro-symbolic learning and reasoning architectures across perception, planning, and control
• Knowledge representations and representation learning for physical AI systems (ontologies, knowledge graphs, scene graphs, procedural and constraint models, neural world models)
• Hybrid approaches combining neural policy learning with symbolic planning and verification
• Methods for safety-aware and constraint-aware decision-making in robotic systems
• Memory and knowledge systems supporting continual learning and experience reuse in embodied agents
• Grounding perceptual observations into structured representations for task-relevant reasoning
• Interpretable and auditable decision processes for robotic systems
• Neuro-symbolic methods for human–robot interaction and collaboration

Submission Guidelines

We suggest 4+N or 8+N paper length formats — i.e., 4 or 8 pages of main content with unlimited additional pages for references, appendices, etc. We impose no strict page length requirements on submissions; we trust that authors will recognize that respecting reviewers' time is helpful to the evaluation of their work.

Our review process will be double-blind.

All accepted papers will be invited for poster presentations; the highest-rated papers, according to the Technical Program Committee, will be given spotlight presentations. Accepted papers will be made available online on this workshop website as non-archival reports, allowing authors to also submit their works to future conferences or journals. We will highlight the Best Paper Award during the closing remarks at the workshop event.

Important Dates

• Submission deadline: TBD, 23:59 AOE.
• Author Notifications: TBD.
• Camera Ready: TBD.
• Workshop: TBD, IROS 2026.

Organizers