Answer Set Programming (ASP) is a declarative formalism widely used in knowledge representation and reasoning for modeling and solving combinatorial problems, yet current Large Language Models (LLMs) often struggle to generate correct programs from natural language specifications. This difficulty stems both from the limited presence of ASP in training corpora and from the strict syntactic and semantic constraints imposed by stable model semantics. We introduce ALM–ASP (Agentic Loop for Modeling in ASP), a multi-agent architecture for automatic ASP modeling grounded in a functional model of language agents equipped with tools and persistent state. ALM–ASP instantiates this model via two interacting agents: a Modeler, which incrementally constructs candidate ASP programs, and a Validator, which assesses their alignment with the original specification and provides feedback for refinement. The agents interact through a shared ASP execution environment backed by the CLINGO engine, yielding an iterative construct–validate loop. An empirical evaluation on a challenging subset of CP–Bench and on problems from recent LP/CP Programming Contests shows that ALM–ASP significantly improves both syntactic validity and end-to-end correctness over general-purpose LLM baselines, and also achieves improved instance coverage compared to the closest agentic alternative, CP–Agent.

Answer Set Programming (ASP) and Large Language Models (LLMs) have emerged as powerful tools in Artificial Intelligence, each offering unique capabilities in knowledge representation and natural language processing, respectively. In this paper, we combine the strengths of the two paradigms with the aim of improving the structured representation of complex knowledge encoded in natural language. In a nutshell, the structured representation is obtained by combining syntactic structures extracted by LLMs and semantic aspects encoded in the knowledge base. The interaction between ASP and LLMs is driven by a YAML file specifying prompt templates and domain-specific background knowledge. The proposed approach is evaluated using a set of benchmarks based on a dataset obtained from problems of ASP Competitions. The results of our experiment show that ASP can sensibly improve the F1-score, especially when relatively small models are used.

ASP Chef enriches Answer Set Programming (ASP) with the notion of recipe, that is, a sequence of operations on answer sets. Recipes are designed and executed in modern browsers, and further improve the fast prototyping capabilities of ASP. This paper introduces new operations designed to integrate Large Language Models (LLMs) in recipe, with the aim of combining the reasoning strength of ASP with the natural language capabilities of LLMs, to enable more interactive and adaptive problem-solving workflows. In a nutshell, answer sets in input are transformed into prompts for LLMs, whose responses are processed to extract facts for subsequent operations within the recipe.

Abstract We present ASP Chef Mustache, an extension of ASP Chef that enhances template-based rendering of answer set programming (ASP) solutions using a logic-less templating system inspired by Mustache. Our approach integrates data visualization frameworks such as Tabulator, Chart.js, and vis.js, enabling interactive representations of ASP interpretations as tables, charts, and graphs. Mustache queries in templates support advanced constructs for formatting, sorting, and multi-stage expansion, facilitating the generation of rich, structured outputs. We demonstrate the power of this framework through a series of use cases, including data analysis for the Italian VQR, visualization of blocking sets in graphs, and scheduling problems. The result is a versatile tool for bridging declarative problem solving and modern web-based visual analytics.

ASP Chef is a versatile tool built upon the principles of Answer Set Programming (ASP), offering a unique approach to problem-solving through the concept of ASP recipes. In this paper, we explore two key components of ASP Chef: the Graph ingredient and one of its extension mechanisms for registering new ingredients. The Graph ingredient serves as a fundamental feature within ASP Chef, allowing users to interpret instances of a designed predicate to construct graphs from the data. Through this capability, ASP Chef facilitates the visualization and analysis of complex relationships and structures inherent in various domains. Furthermore, ASP Chef offers a flexible extension mechanism that empowers users to register new recipes as custom ingredients. These custom ingredients, defined by sequences of mappings from interpretations to interpretations, can be stored locally within the local storage of the browser. This enables users to expand the capabilities of ASP Chef to suit their specific needs and use cases, fostering a collaborative environment where users can share and reuse custom ingredients seamlessly. Notably, the addition of new ingredients does not impose requirements on the utilization of recipes that employ them, underscoring the modular and interoperable design of ASP Chef.