nl2spec: Interactively Translating Unstructured Natural Language to Temporal Logics with Large Language Models

Abstract

A rigorous formalization of desired system requirements is indispensable when performing any verification task. This often limits the application of verification techniques, as writing formal specifications is an error-prone and time-consuming manual task. To facilitate this, we present , a framework for applying Large Language Models (LLMs) to derive formal specifications (in temporal logics) from unstructured natural language. In particular, we introduce a new methodology to detect and resolve the inherent ambiguity of system requirements in natural language: we utilize LLMs to map subformulas of the formalization back to the corresponding natural language fragments of the input. Users iteratively add, delete, and edit these sub-translations to amend erroneous formalizations, which is easier than manually redrafting the entire formalization. The framework is agnostic to specific application domains and can be extended to similar specification languages and new neural models. We perform a user study to obtain a challenging dataset, which we use to run experiments on the quality of translations. We provide an open-source implementation, including a web-based frontend.

NeuroSynt: A Neuro-symbolic Portfolio Solver for Reactive Synthesis

“…In contrast to proof guidance, LLMs can be used for end-to-end generation and repair of proofs in Isabelle/HOL [31]. LLMs have recently also enabled a step towards autoformalization of unstructured natural language for theorem proving [37,64] and temporal logic [19]. Further, deep learning has had a considerable impact on program verification and synthesis, i.e., for termination analysis [3,32], creating loop invariants [49,56,61] and program synthesis/induction [2,18,26,28].…”