Why This Matters

Traditional planning algorithms like Monte Carlo Tree Search achieve strong performance but lack transparency, making their outputs unsuitable for high-stakes CPS applications where users need to understand and trust system decisions. LogiEx is innovative because it bridges the transparency gap by combining stochastic search with formal logic verification, allowing the system to explain not just what actions it recommends but why those actions are justified given domain constraints and objectives. This addresses a critical safety concern in AI-driven CPS.

What We Did

LogiEx integrates formal logic and large language models to provide explainable sequential planning for human-centered cyber-physical systems like intelligent transportation. The system combines Monte Carlo Tree Search planning with logical reasoning to generate trustworthy explanations for planning decisions. LogiEx categorizes user queries into three types: those answerable from existing search trees, those requiring human-guided search, and those requiring background knowledge. The framework generates logical evidence supporting planning decisions and translates this into natural language explanations that users can verify.

Key Results

Quantitative evaluation demonstrates LogiEx achieves up to 7.9x higher semantic similarity and 1.7x higher factual consistency compared to LLM-only baselines on explaining transportation planning decisions. User studies validate that the framework provides faithful, consistent explanations that help users understand the planning process while maintaining the ability to ask follow-up questions for deeper reasoning.

Full Abstract

Cite This Paper

@inproceedings{iccps2026_logiex,
  author = {An, Ziyan and Wang, Xia and Baier, Hendrik and Chen, Zirong and Dubey, Abhishek and Mukhopadhyay, Ayan and Johnson, Taylor T. and Sprinkle, Jonathan and Ma, Meiyi},
  title = {LogiEx: Logic-Integrated Explanations for Stochastic Planning in Cyber-Physical Systems},
  year = {2026},
  booktitle = {Proceedings of the HSCC/ICCPS 2026: 29th ACM International Conference on Hybrid Systems: Computation and Control and 17th ACM/IEEE International Conference on Cyber-Physical Systems},
  location = {Saint Malo, France},
  abstract = {Human-centered cyber-physical systems (CPS), such as intelligent transportation services, warehouse robotics operated by human supervisors, and healthcare infrastructures involving clinicians and medical staff, increasingly rely on Artificial Intelligence (AI)-driven sequential decision-making under uncertainty. However, the lack of transparent reasoning in these systems limits trust, verifiability, and human oversight. This challenge is particularly acute for planning algorithms like Monte Carlo Tree Search (MCTS), whose stochastic search processes are opaque to engineers and operators. To address this gap, we introduce LogiEx, a logic-integrated framework that combines large language models (LLMs) with formal methods to generate trustworthy explanations for planning behavior. LogiEx transforms free-form user queries into logical statements with templated variables, then verifies whether evidence extracted from the decision process aligns with both the environment state and the constraints of the stochastic planning model. This enables grounded explanations across a wide range of user questions—from factual retrieval to comparative reasoning. LogiEx also supports Human-Guided Search (HuGS), allowing users to pose conditional `what-if'' queries that trigger new, scenario-specific searches, ensuring that humans are not passive observers but active participants who can steer and refine the planning process. We evaluate LogiEx through both quantitative assessments and user studies, finding that it consistently outperforms baselines, achieving up to 7.9 higher semantic similarity (BERTScore) and 1.6 higher factual consistency (FactCC) compared to baseline LLMs, and is the most preferred form of explanation among CPS practitioners.},
  keywords = {explainable AI, transit planning, formal logic, large language models, Monte Carlo tree search, knowledge graphs, human-AI interaction},
  note = {Acceptance rate: 28\%; Regular Paper; Track: Systems and Applications},
  series = {HSCC/ICCPS '26}
}
Quick Info
Year 2026
Series HSCC/ICCPS '26
Keywords
explainable AI transit planning formal logic large language models Monte Carlo tree search knowledge graphs human-AI interaction
Research Areas
transit planning Explainable AI
Search Tags

LogiEx, Logic, Integrated, Explanations, Stochastic, Planning, Cyber, Physical, Systems, explainable AI, transit planning, formal logic, large language models, Monte Carlo tree search, knowledge graphs, human-AI interaction, transit, planning, Explainable AI, 2026, An, Wang, Baier, Chen, Dubey, Mukhopadhyay, Johnson, Sprinkle, Ma, HSCC/ICCPS26