About The Workshop
The 1st Workshop on Cognitive Architectures for Robotics: LLMs and Logic in Action (CARLA) seeks to transform the landscape of intelligent robotics by pioneering the integration of large language models (LLMs), symbolic reasoning, and logic solvers into robotic systems. As robotics moves towards real-world applications requiring adaptability, safety, and complex decision-making, this workshop focuses on harnessing the synergy between data-driven learning models and symbolic logic-based systems to advance automation.
CARLA highlights the complementary roles of LLMs and traditional symbolic solvers in advancing robotic reasoning and control. While LLMs serve as cognitive engines, excelling in processing natural language and enabling dynamic adaptation to changing environments, their statistical nature makes them less suitable for precise logical inferences and planning. For these tasks, traditional symbolic solvers offer a robust and reliable foundation. By integrating the strengths of LLMs with the precision of symbolic reasoning and planning, this workshop aims to foster the development of intelligent, reliable, and explainable robotic systems capable of seamless interaction with humans and their surroundings.
Schedule
| Time (SGT) | Session | Speaker | Talk Title | Host |
|---|---|---|---|---|
| 08:30 – 08:40 | Introduction and Opening Remarks | N/A |
Topics
The workshop will cover a range of topics, including but not limited to:
- Knowledge Representation for Robotics:
Frameworks and methodologies for integrating structured knowledge into robotic architectures. - LLMs as Cognitive Engines:
Leveraging LLMs to process complex commands, generate actionable insights, and facilitate human-robot communication. - Neuro-Symbolic Systems in Robotics:
Hybrid approaches combining neural networks with symbolic reasoning for decision-making and task execution. - Logic Solvers in Robotic Control:
Employing constraint solvers, SAT solvers, or theorem provers for autonomous planning and reliable decision-making. - Adaptive and Safe Robotic Architectures:
Strategies for real-time adaptation and coordination in robotic systems using LLMs and symbolic logic. - Human-Robot Collaboration:
Innovations in language-driven interactions, focusing on usability, interpretability, and reliability in diverse scenarios. - Challenges of LLM Deployment in Robotics:
Addressing safety, robustness, and ethical considerations when integrating LLMs into robotic systems. - LLM-Driven Planning and Problem Solving:
Enabling robots to perform dynamic planning and adapt to novel situations using LLM-guided reasoning. - Agentic AI in Robotics:
Designing interaction frameworks for autonomous goal-directed agents through the integration of LLMs with logic control.
Call For Papers
Categories
- Original Research Papers: Presenting novel research contributions aligned with the workshop's focus areas.
- Work-in-Progress Reports: Sharing preliminary findings and ongoing research efforts.
- Position Papers: Discussing emerging challenges, visionary ideas, and future directions at the intersection of robotics, LLMs, and logic solvers.
- Already Presented Works: Contributions that have been previously presented at other venues but are relevant to the workshop themes. Such works can be resubmitted to foster further discussion and exploration.
Important Dates
- Paper Deadline: June 1, 2025 (AOE)
- Notification: July 13, 2025 (AOE)
- Camera-ready: July 27, 2025 (AOE)
Submission Guidelines
Formatting Requirements
Submissions must be in English and follow the CEURART style: CEURART Paper Style.Papers must be submitted as a single PDF file:
- Maximum of 9 pages (excluding references) for full papers and 4 pages (excluding references) for short papers.
- All submissions must be in English and submitted in PDF format.
Submissions exceeding the page limit will be desk rejected.
Submission Site
Submissions will be managed via Microsoft CMT. Papers will remain private during the review process. All authors must maintain up-to-date Microsoft CMT profiles to ensure proper conflict-of-interest management and paper matching. Incomplete profiles may result in desk rejection.
Submit papers through the C.A.R.L.A. Submission Portal on Microsoft CMT (C.A.R.L.A. Workshop Submission Portal).
Anonymity
The workshop follows a single-blind review process. Submissions must not be anonymized by removing author names, affiliations, and acknowledgments.Dual Submission and Non-Archival Policy
Submissions under review at other venues will be accepted, provided they do not breach any dual-submission or anonymity policies of those venues. Submissions will not be indexed or have archival proceedings.Transparency
By submitting to the C.A.R.L.A. Workshop, authors agree that for all accepted papers, the original submission, reviews, and meta-reviews will be made publicly available on OpenReview.Contact
Email at: carla.ws.info@gmail.comAccepted Paper
Speakers and Panelists (Tentative)
Organizers
This workshop is organized by
Sponsors
We thank our sponsors for their generous support!