13 citations · 26 across the 9 of their papers we have counts for
9 papers
Natural Synthesis: Outperforming Reactive Synthesis Tools with Large Reasoning Models
Frederik Schmitt, Matthias Cosler, Niklas Metzger +4
Reactive synthesis, the problem of automatically constructing a hardware circuit from a logical specification, is a long-standing challenge in formal verification. It is elusive fo…
Learning Better Representations From Less Data For Propositional Satisfiability
Mohamed Ghanem, Frederik Schmitt, Julian Siber +1
Training neural networks on NP-complete problems typically demands very large amounts of training data and often needs to be coupled with computationally expensive symbolic verifie…
NeuroSynt: A Neuro-symbolic Portfolio Solver for Reactive Synthesis
Matthias Cosler, Christopher Hahn, Ayham Omar +1
We introduce NeuroSynt, a neuro-symbolic portfolio solver framework for reactive synthesis. At the core of the solver lies a seamless integration of neural and symbolic approaches…
nl2spec: Interactively Translating Unstructured Natural Language to Temporal Logics with Large Language Models
Matthias Cosler, Christopher Hahn, Daniel Mendoza +2
A rigorous formalization of desired system requirements is indispensable when performing any verification task. This often limits the application of verification techniques, as wri…
Iterative Circuit Repair Against Formal Specifications
Matthias Cosler, Frederik Schmitt, Christopher Hahn +1
We present a deep learning approach for repairing sequential circuits against formal specifications given in linear-time temporal logic (LTL). Given a defective circuit and its for…
Formal Specifications from Natural Language
Christopher Hahn, Frederik Schmitt, Julia J. Tillman +3
We study the generalization abilities of language models when translating natural language into formal specifications with complex semantics. In particular, we fine-tune language m…
Attention Flows for General Transformers
Niklas Metzger, Christopher Hahn, Julian Siber +2
In this paper, we study the computation of how much an input token in a Transformer model influences its prediction. We formalize a method to construct a flow network out of the at…
Neural Circuit Synthesis from Specification Patterns
Frederik Schmitt, Christopher Hahn, Markus N. Rabe +1
We train hierarchical Transformers on the task of synthesizing hardware circuits directly out of high-level logical specifications in linear-time temporal logic (LTL). The LTL synt…
Teaching Temporal Logics to Neural Networks
Christopher Hahn, Frederik Schmitt, Jens U. Kreber +2
We study two fundamental questions in neuro-symbolic computing: can deep learning tackle challenging problems in logics end-to-end, and can neural networks learn the semantics of l…