Vı́ctor Braberman

Tomás Delgado, Marco Sánchez Sorondo, Víctor Braberman et al.

Controller synthesis is in essence a case of model-based planning for non-deterministic environments in which plans (actually ''strategies'') are meant to preserve system goals indefinitely. In the case of supervisory control environments are specified as the parallel composition of state machines and valid strategies are required to be ''non-blocking'' (i.e., always enabling the environment to reach certain marked states) in addition to safe (i.e., keep the system within a safe zone). Recently, On-the-fly Directed Controller Synthesis techniques were proposed to avoid the exploration of the entire -and exponentially large-environment space, at the cost of non-maximal permissiveness, to either find a strategy or conclude that there is none. The incremental exploration of the plant is currently guided by a domain-independent human-designed heuristic. In this work, we propose a new method for obtaining heuristics based on Reinforcement Learning (RL). The synthesis algorithm is thus framed as an RL task with an unbounded action space and a modified version of DQN is used. With a simple and general set of features that abstracts both states and actions, we show that it is possible to learn heuristics on small versions of a problem that generalize to the larger instances, effectively doing zero-shot policy transfer. Our agents learn from scratch in a highly partially observable RL task and outperform the existing heuristic overall, in instances unseen during training.

Víctor A. Braberman, Flavia Bonomo-Braberman, Yiannis Charalambous et al.

The emergence of prompting as the dominant paradigm for leveraging Large Language Models (LLMs) has led to a proliferation of LLM-native software, where application behavior arises from complex, stochastic data transformations. However, the engineering of such systems remains largely exploratory and ad-hoc, hampered by the absence of conceptual frameworks, ex-ante methodologies, design guidelines, and specialized benchmarks. We argue that a foundational step towards a more disciplined engineering practice is a systematic understanding of the core functional units--generative transformations--and their compositional patterns within LLM-native applications. Focusing on the rich domain of software verification and falsification, we conduct a secondary study of over 100 research proposals to address this gap. We first present a fine-grained taxonomy of generative transformations, abstracting prompt-based interactions into conceptual signatures. This taxonomy serves as a scaffolding to identify recurrent transformation relationship patterns--analogous to software design patterns--that characterize solution approaches in the literature. Our analysis not only validates the utility of the taxonomy but also surfaces strategic gaps and cross-dimensional relationships, offering a structured foundation for future research in modular and compositional LLM application design, benchmarking, and the development of reliable LLM-native systems.

Rodrigo Castaño, Victor Braberman, Diego Garbervetsky et al.

Software Model Checkers have shown outstanding performance improvements in recent times. Moreover, for specific use cases, formal verification techniques have shown to be highly effective, leading to a number of high-profile success stories. However, widespread adoption remains unlikely in the short term and one of the remaining obstacles in that direction is the vast number of instances which software model checkers cannot fully analyze within reasonable memory and CPU bounds. The majority of verification tools fail to provide a measure of progress or any intermediate verification result when such situations occur. Inspired in the success that coverage metrics have achieved in industry, we propose to adapt the definition of coverage to the context of verification. We discuss some of the challenges in pinning down a definition that resembles the deeply rooted semantics of test coverage. Subsequently we propose a definition for a broad family of verification techniques: those based on Abstract Reachability Trees. Moreover, we discuss a general approach to computing an under-approximation of such metric and a specific heuristic to improve the performance. Finally, we conduct an empirical evaluation to assess the viability of our approach.

Rodrigo Castaño, Victor Braberman, Diego Garbervetsky et al.

Software model checking constitutes an undecidable problem and, as such, even an ideal tool will in some cases fail to give a conclusive answer. In practice, software model checkers fail often and usually do not provide any information on what was effectively checked. The purpose of this work is to provide a conceptual framing to extend software model checkers in a way that allows users to access information about incomplete checks. We characterize the information that model checkers themselves can provide, in terms of analyzed traces, i.e. sequences of statements, and safe cones, and present the notion of execution reports, which we also formalize. We instantiate these concepts for a family of techniques based on Abstract Reachability Trees and implement the approach using the software model checker CPAchecker. We evaluate our approach empirically and provide examples to illustrate the execution reports produced and the information that can be extracted.

Daniel Ciolek, Victor Braberman, Nicolás D'Ippolito et al.

This paper presents a Directed Controller Synthesis (DCS) technique for discrete event systems. The DCS method explores the solution space for reactive controllers guided by a domain-independent heuristic. The heuristic is derived from an efficient abstraction of the environment based on the componentized way in which complex environments are described. Then by building the composition of the components on-the-fly DCS obtains a solution by exploring a reduced portion of the state space. This work focuses on untimed discrete event systems with safety and co-safety (i.e. reachability) goals. An evaluation for the technique is presented comparing it to other well-known approaches to controller synthesis (based on symbolic representation and compositional analyses).