SemML 2.0: Synthesizing Controllers for LTL

2026-04-27

AI Summary

This work addresses the problem of automatically synthesizing reactive controllers from Linear Temporal Logic (LTL) specifications for safety-critical systems. We propose a novel framework that integrates automata-theoretic techniques, partial state-space exploration, and machine learning–guided search, enhanced with multiple heuristic strategies and optimized using AIGER/Mealy representations. This approach achieves a breakthrough balance between synthesis efficiency and controller compactness. Evaluated on the SYNTCOMP benchmark suite, our method substantially outperforms state-of-the-art tools—including Strix, LtlSynt, and SemML 1.0—by solving more instances faster while maintaining industry-leading solution quality.

Technology Category

• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Planning, Routing, and Scheduling: Learning for Planning and Scheduling

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Synthesizing a reactive system from specifications given in linear temporal logic (LTL) is a classical problem, finding its applications in safety-critical systems design. These systems are typically represented using either Mealy machines or AIGER circuits. We present the second version of SemML, which outperforms all state-of-the-art tools for finding either solution. Aside from implementing the classical automata-theoretic approach, our tool utilizes partial exploration and machine-learning guidance for obtaining solutions efficiently, and numerous heuristics and improvements of classic algorithms for extracting small representations of these solutions. We evaluate our tool against the existing state-of-the-art tools (in particular Strix, LtlSynt, and the previous version of SemML) on the dataset of the synthesis competition SYNTCOMP. We show that we solve significantly more instances and do so much faster than other tools, while maintaining state-of-the-art solution quality.

Problem

• controller synthesis
• linear temporal logic
• reactive systems
• safety-critical systems

Innovation

• LTL synthesis
• machine learning guidance
• partial exploration
• Mealy machines
• AIGER circuits