arXiv subjects

Federico Mora

Publications and source records attributed to Federico Mora.

10 recordsLinked to original sources

arXiv2026

Learning Context-Free Grammars for Grammar-Constrained Decoding via Declarative Agentic Programming with Guarantees

Language models (LMs) are increasingly used to interact with external services via programs written in domain-specific languages (DSLs). Unfortunately, since DSLs are often low-resource and esoteric, LMs frequently produce syntactically invalid programs in these languages. Grammar-constrained decoding can eliminate such failures, but requires syntactic constraints. These are usually in the form of a context-free grammar for the target language, an artifact that is hard to come by for third-party DSLs. In this work, we define an agent, called Autogrammar, that automatically learns context-free grammars from documentation and execution data. Autogrammar is formalized as a Kripke structure whose nondeterministic choices are resolved by a language model, enabling declarative control of agent behavior via linear temporal logic constraints. We evaluate four versions of Autogrammar on three DSLs (i.e., Amazon CloudWatch Logs Insights, Dynatrace Query Language, and Datadog Search Syntax) and find that it generates grammars that achieve near perfect precision on unseen data; that temporal restrictions reduce execution time by 3.8x without incurring statistically-significant loss in precision; that execution data is crucial while documentation is dispensable; and that grammar-constrained decoding using Autogrammar-generated grammars significantly improves end-to-end LM performance on eight out of ten real tasks, matching or exceeding the performance of a professionally-maintained grammar. In comparison, the context-free grammars generated by existing LM baselines and a state-of-the-art formal technique perform significantly worse over the same evaluation.

cs.PL↗

arXiv2025

Learning Affordances at Inference-Time for Vision-Language-Action Models

Solving complex real-world control tasks often takes multiple tries: if we fail at first, we reflect on what went wrong, and change our strategy accordingly to avoid making the same mistake. In robotics, Vision-Language-Action models (VLAs) offer a promising path towards solving complex control tasks, but lack the ability to contextually and dynamically readjust behavior when they fail to accomplish a task. In this work, we introduce Learning from Inference-Time Execution (LITEN), which connects a VLA low-level policy to a high-level VLM that conditions on past experiences by including them in-context, allowing it to learn the affordances and capabilities of the low-level VLA. Our approach iterates between a reasoning phase that generates and executes plans for the low-level VLA, and an assessment phase that reflects on the resulting execution and draws useful conclusions to be included in future reasoning contexts. Unlike similar approaches to self-refinement in non-robotics domains, LITEN must reflect on unstructured real-world robot trajectories (e.g., raw videos), which requires structured guiderails during assessment. Our experimental results demonstrate LITEN is able to effectively learn from past experience to generate plans that use high-affordance instructions to accomplish long-horizon tasks.

cs.RO↗

arXiv2025

Online Prompt Selection for Program Synthesis

Large Language Models (LLMs) demonstrate impressive capabilities in the domain of program synthesis. This level of performance is not, however, universal across all tasks, all LLMs and all prompting styles. There are many areas where one LLM dominates, one prompting style dominates, or where calling a symbolic solver is a better choice than an LLM. A key challenge for the user then, is to identify not only when an LLM is the right choice of solver, and the appropriate LLM to call for a given synthesis task, but also the right way to call it. A non-expert user who makes the wrong choice, incurs a cost both in terms of results (number of tasks solved, and the time it takes to solve them) and financial cost, if using a closed-source language model via a commercial API. We frame this choice as an online learning problem. We use a multi-armed bandit algorithm to select which symbolic solver, or LLM and prompt combination to deploy in order to maximize a given reward function (which may prioritize solving time, number of synthesis tasks solved, or financial cost of solving). We implement an instance of this approach, called CYANEA, and evaluate it on synthesis queries from the literature in ranking function synthesis, from the syntax-guided synthesis competition, and fresh, unseen queries generated from SMT problems. CYANEA solves 37.2% more queries than the best single solver and achieves results within 4% of the virtual best solver.

cs.AI↗

arXiv2024

Synthetic Programming Elicitation for Text-to-Code in Very Low-Resource Programming and Formal Languages

Recent advances in large language models (LLMs) for code applications have demonstrated remarkable zero-shot fluency and instruction following on challenging code related tasks ranging from test case generation to self-repair. Unsurprisingly, however, models struggle to compose syntactically valid programs in programming languages unrepresented in pre-training, referred to as very low-resource Programming Languages (VLPLs). VLPLs appear in crucial settings, including domain-specific languages for internal tools, tool-chains for legacy languages, and formal verification frameworks. Inspired by a technique called natural programming elicitation, we propose designing an intermediate language that LLMs "naturally" know how to use and which can be automatically compiled to a target VLPL. When LLMs generate code that lies outside of this intermediate language, we use compiler techniques to repair the code into programs in the intermediate language. Overall, we introduce \emph{synthetic programming elicitation and compilation} (SPEAC), an approach that enables LLMs to generate syntactically valid code even for VLPLs. We empirically evaluate the performance of SPEAC in a case study for the UCLID5 formal verification language and find that, compared to existing retrieval and fine-tuning baselines, SPEAC produces syntactically correct programs more frequently and without sacrificing semantic correctness.

cs.PL↗

arXiv2023

An Eager Satisfiability Modulo Theories Solver for Algebraic Datatypes

Algebraic data types (ADTs) are a construct classically found in functional programming languages that capture data structures like enumerated types, lists, and trees. In recent years, interest in ADTs has increased. For example, popular programming languages, like Python, have added support for ADTs. Automated reasoning about ADTs can be done using satisfiability modulo theories (SMT) solving, an extension of the Boolean satisfiability problem with constraints over first-order structures. Unfortunately, SMT solvers that support ADTs do not scale as state-of-the-art approaches all use variations of the same \emph{lazy} approach. In this paper, we present an SMT solver that takes a fundamentally different approach, an \emph{eager} approach. Specifically, our solver reduces ADT queries to a simpler logical theory, uninterpreted functions (UF), and then uses an existing solver on the reduced query. We prove the soundness and completeness of our approach and demonstrate that it outperforms the state-of-theart on existing benchmarks, as well as a new, more challenging benchmark set from the planning domain.

cs.LO↗

arXiv2023

Genetic Algorithms for Searching a Matrix of Metagrammars for Synthesis

Syntax-guided synthesis is a paradigm in program synthesis in which the search space of candidate solutions is constrained by a syntactic template in the form of a grammar. These syntactic constraints serve two purposes: constraining the language to the space the user desires, but also rendering the search space tractable for the synthesizer. Given a well-written syntactic template, this is an extremely effective technique. However, this is highly dependent on the user providing such a template: a syntactic template that is too large results in a larger search space and slower synthesis, and a syntactic template that is too small may not contain the solution needed. In this work, we frame the space of syntactic templates as a matrix of rules, and demonstrate how this matrix can be searched effectively with little training data using simple search techniques such as genetic algorithms, giving improvements in both the number of benchmarks solved and solving time for the state-of-the-art synthesis solver.

cs.SE↗

arXiv2022

UCLID5: Multi-Modal Formal Modeling, Verification, and Synthesis

UCLID5 is a tool for the multi-modal formal modeling, verification, and synthesis of systems. It enables one to tackle verification problems for heterogeneous systems such as combinations of hardware and software, or those that have multiple, varied specifications, or systems that require hybrid modes of modeling. A novel aspect of \uclid is an emphasis on the use of syntax-guided and inductive synthesis to automate steps in modeling and verification. This tool paper presents new developments in the \uclid tool including new language features, integration with new techniques for syntax-guided synthesis and satisfiability solving, support for hyperproperties and combinations of axiomatic and operational modeling, demonstrations on new problem classes, and a robust implementation.

cs.LO↗

arXiv2021

String Theories involving Regular Membership Predicates: From Practice to Theory and Back

Widespread use of string solvers in formal analysis of string-heavy programs has led to a growing demand for more efficient and reliable techniques which can be applied in this context, especially for real-world cases. Designing an algorithm for the (generally undecidable) satisfiability problem for systems of string constraints requires a thorough understanding of the structure of constraints present in the targeted cases. In this paper, we investigate benchmarks presented in the literature containing regular expression membership predicates, extract different first order logic theories, and prove their decidability, resp. undecidability. Notably, the most common theories in real-world benchmarks are PSPACE-complete and directly lead to the implementation of a more efficient algorithm to solving string constraints.

cs.CL↗

arXiv2021

An SMT Solver for Regular Expressions and Linear Arithmetic over String Length

We present a novel length-aware solving algorithm for the quantifier-free first-order theory over regex membership predicate and linear arithmetic over string length. We implement and evaluate this algorithm and related heuristics in the Z3 theorem prover. A crucial insight that underpins our algorithm is that real-world instances contain a wealth of information about upper and lower bounds on lengths of strings under constraints, and such information can be used very effectively to simplify operations on automata representing regular expressions. Additionally, we present a number of novel general heuristics, such as the prefix/suffix method, that can be used in conjunction with a variety of regex solving algorithms, making them more efficient. We showcase the power of our algorithm and heuristics via an extensive empirical evaluation over a large and diverse benchmark of 57256 regex-heavy instances, almost 75% of which are derived from industrial applications or contributed by other solver developers. Our solver outperforms five other state-of-the-art string solvers, namely, CVC4, OSTRICH, Z3seq, Z3str3, and Z3-Trau, over this benchmark, in particular achieving a 2.4x speedup over CVC4, 4.4x speedup over Z3seq, 6.4x speedup over Z3-Trau, 9.1x speedup over Z3str3, and 13x speedup over OSTRICH.

cs.LO↗

arXiv2020

Synthesis in Uclid5

We describe an integration of program synthesis into Uclid5, a formal modelling and verification tool. To the best of our knowledge, the new version of Uclid5 is the only tool that supports program synthesis with bounded model checking, k-induction, sequential program verification, and hyperproperty verification. We use the integration to generate 25 program synthesis benchmarks with simple, known solutions that are out of reach of current synthesis engines, and we release the benchmarks to the community.

cs.PL↗