arXiv subjects

Claudio Sacerdoti Coen

Publications and source records attributed to Claudio Sacerdoti Coen.

At least 19 recordsLinked to original sources

arXiv2026

A Reversible Crumbling Abstract Machine for Plotkin's Call-by-Value

Landauer's embeddings enable the reversibility of computations for non-reversible programming languages, augmenting each intermediate state with enough data to reconstruct the previous state. An interesting research question is therefore to try to reduce the space overhead required. In this work we propose a Landauer's embedding for Plotkin's call-by-value calculus (CbV). In order to control the computational complexity of CbV and turn the number of $β$-steps into a cost model, CbV is typically implemented via reduction machines. We show that one machine, that has not received much attention, exhibits a particularly compact Landauer's embedding, requiring only constant space overhead for each step.

cs.LO↗

arXiv2026

Thinking Machines: Mathematical Reasoning in the Age of LLMs

Large Language Models (LLMs) have demonstrated impressive capabilities in structured reasoning and symbolic tasks, with coding emerging as a particularly successful application. This progress has naturally motivated efforts to extend these models to mathematics, both in its traditional form, expressed through natural-style mathematical language, and in its formalized counterpart, expressed in a symbolic syntax suitable for automatic verification. Yet, despite apparent parallels between programming and proof construction, advances in formalized mathematics have proven significantly more challenging. This gap raises fundamental questions about the nature of reasoning in current LLM architectures, the role of supervision and feedback, and the extent to which such models maintain an internal notion of computational or deductive state. In this article, we review the current state-of-the-art in mathematical reasoning with LLMs, focusing on recent models and benchmarks. We explore three central issues at the intersection of machine learning and mathematical cognition: (i) the trade-offs between traditional and formalized mathematics as training and evaluation domains; (ii) the structural and methodological reasons why proof synthesis remains more brittle than code generation; and (iii) whether LLMs genuinely represent or merely emulate a notion of evolving logical state. Our goal is not to draw rigid distinctions but to clarify the present boundaries of these systems and outline promising directions for their extension.

cs.AI↗

arXiv2025

The Cost of Skeletal Call-by-Need, Smoothly

Skeletal call-by-need is an optimization of call-by-need evaluation also known as "fully lazy sharing": when the duplication of a value has to take place, it is first split into "skeleton", which is then duplicated, and "flesh" which is instead kept shared. Here, we provide two cost analyses of skeletal call-by-need. Firstly, we provide a family of terms showing that skeletal call-by-need can be asymptotically exponentially faster than call-by-need in both time and space; it is the first such evidence, to our knowledge. Secondly, we prove that skeletal call-by-need can be implemented efficiently, that is, with bi-linear overhead. This result is obtained by providing a new smooth presentation of ideas by Shivers and Wand for the reconstruction of skeletons, which is then smoothly plugged into the study of an abstract machine following the distillation technique by Accattoli et al.

cs.LO↗

arXiv2025

Positive Sharing and Abstract Machines

Wu's positive $λ$-calculus is a recent call-by-value $λ$-calculus with sharing coming from Miller and Wu's study of the proof-theoretical concept of focalization. Accattoli and Wu showed that it simplifies a technical aspect of the study of sharing; namely it rules out the recurrent issue of renaming chains, that often causes a quadratic time slowdown. In this paper, we define the natural abstract machine for the positive $λ$-calculus and show that it suffers from an inefficiency: the quadratic slowdown somehow reappears when analyzing the cost of the machine. We then design an optimized machine for the positive $λ$-calculus, which we prove efficient. The optimization is based on a new slicing technique which is dual to the standard structure of machine environments.

cs.LO↗

arXiv2025

Closure Conversion, Flat Environments, and the Complexity of Abstract Machines

Closure conversion is a program transformation at work in compilers for functional languages to turn inner functions into global ones, by building closures pairing the transformed functions with the environment of their free variables. Abstract machines rely on similar and yet different concepts of closures and environments. In this paper, we study the relationship between the two approaches. We adopt a very simple λ-calculus with tuples as source language and study abstract machines for both the source language and the target of closure conversion. Moreover, we focus on the simple case of flat closures/environments, that is, with no sharing of environments. We provide three contributions. Firstly, a new simple proof technique for the correctness of closure conversion, inspired by abstract machines. Secondly, we show how the closure invariants of the target language allow us to design a new way of handling environments in abstract machines, not suffering the shortcomings of other styles. Thirdly, we study the machines from the point of view of time complexity, adapting analyses by Accattoli and co-authors. We show that closure conversion decreases various dynamic costs while increasing the size of the initial code. Despite these changes, the overall complexity of the machines before and after closure conversion turns out to be the same.

cs.PL↗

arXiv2024

Proceedings Workshop on Logical Frameworks and Meta-Languages: Theory and Practice

Logical frameworks and meta-languages form a common substrate for representing, implementing and reasoning about a wide variety of deductive systems of interest in logic and computer science. Their design, implementation and their use in reasoning tasks, ranging from the correctness of software to the properties of formal systems, have been the focus of considerable research over the last three decades. The LFMTP workshop brought together designers, implementors and practitioners to discuss various aspects impinging on the structure and utility of logical frameworks, including the treatment of variable binding, inductive and co-inductive reasoning techniques and the expressiveness and lucidity of the reasoning process. The 2024 instance of LFMTP was organized by Florian Rabe and Claudio Sacerdoti Coen in Tallinn, Estonia, the 8th July, as a satellite event of the FSCD conference. The workshop received 8 submissions, of which 6 were presented at the workshop. Of these, 2 were work-in-progress presentations, and 4 were accepted for these formal proceedings. Additionally, Carsten Schürmann of IT University of Copenhagen gave an invited talk on Nominal State Separating Proofs.

cs.LO↗

arXiv2024

IMELL Cut Elimination with Linear Overhead

Recently, Accattoli introduced the Exponential Substitution Calculus (ESC) given by untyped proof terms for Intuitionistic Multiplicative Exponential Linear Logic (IMELL), endowed with rewriting rules at-a-distance for cut elimination. He also introduced a new cut elimination strategy, dubbed the good strategy, and showed that its number of steps is a time cost model with polynomial overhead for the ESC/IMELL, and the first such one. Here, we refine Accattoli's result by introducing an abstract machine for ESC and proving that it implements the good strategy and computes cut-free terms/proofs within a linear overhead.

cs.LO↗

arXiv2021

Strong Call-by-Value is Reasonable, Implosively

Whether the number of beta-steps in the lambda-calculus can be taken as a reasonable time cost model (that is, polynomially related to the one of Turing machines) is a delicate problem, which depends on the notion of evaluation strategy. Since the nineties, it is known that weak (that is, out of abstractions) call-by-value evaluation is a reasonable strategy while Lévy's optimal parallel strategy, which is strong (that is, it reduces everywhere), is not. The strong case turned out to be subtler than the weak one. In 2014 Accattoli and Dal Lago have shown that strong call-by-name is reasonable, by introducing a new form of useful sharing and, later, an abstract machine with an overhead quadratic in the number of beta-steps. Here we show that also strong call-by-value evaluation is reasonable for time, via a new abstract machine realizing useful sharing and having a linear overhead. Moreover, our machine uses a new mix of sharing techniques, adding on top of useful sharing a form of implosive sharing, which on some terms brings an exponential speed-up. We give examples of families that the machine executes in time logarithmic in the number of beta-steps.

cs.LO↗

arXiv2021

Proceedings Fifteenth Workshop on Logical Frameworks and Meta-Languages: Theory and Practice

This volume contains a selection of papers presented at LFMTP 2020, the 15th International Workshop on Logical Frameworks and Meta-Languages: Theory and Practice (LFMTP), held the 29-30th of June, 2019, using the Zoom video conferencing tool due to COVID restrictions. Officially the workshop was held in Paris, France, and it was affiliated with IJCAR 2020, FSCD 2020 and many other satellite events. Logical frameworks and meta-languages form a common substrate for representing, implementing and reasoning about a wide variety of deductive systems of interest in logic and computer science. Their design, implementation and their use in reasoning tasks, ranging from the correctness of software to the properties of formal systems, have been the focus of considerable research over the last two decades. This workshop will bring together designers, implementors and practitioners to discuss various aspects impinging on the structure and utility of logical frameworks, including the treatment of variable binding, inductive and co-inductive reasoning techniques and the expressiveness and lucidity of the reasoning process.

cs.LO↗

arXiv2020

Pre-Proceedings of the 28th International Workshop on Functional and Logic Programming (WFLP 2020)

This volume constitutes the pre-proceedings of the 28th International Workshop on Functional and Logic Programming (WFLP 2020), organized by the University of Bologna, Italy, as part of Bologna Federated Conference on Programming Languages 2020. The international Workshop on Functional and (constraint) Logic Programming (WFLP) aims at bringing together researchers, students, and practitioners interested in functional programming, logic programming, and their integration. WFLP has a reputation for being a lively and friendly forum, and it is open for presenting and discussing work in progress, technical contributions, experience reports, experiments, reviews, and system descriptions.

cs.PL↗

arXiv2019

Crumbling Abstract Machines

Extending the lambda-calculus with a construct for sharing, such as let expressions, enables a special representation of terms: iterated applications are decomposed by introducing sharing points in between any two of them, reducing to the case where applications have only values as immediate subterms. This work studies how such a crumbled representation of terms impacts on the design and the efficiency of abstract machines for call-by-value evaluation. About the design, it removes the need for data structures encoding the evaluation context, such as the applicative stack and the dump, that get encoded in the environment. About efficiency, we show that there is no slowdown, clarifying in particular a point raised by Kennedy, about the potential inefficiency of such a representation. Moreover, we prove that everything smoothly scales up to the delicate case of open terms, needed to implement proof assistants. Along the way, we also point out that continuation-passing style transformations--that may be alternatives to our representation--do not scale up to the open case.

cs.LO↗

arXiv2019

Sharing Equality is Linear

The $λ$-calculus is a handy formalism to specify the evaluation of higher-order programs. It is not very handy, however, when one interprets the specification as an execution mechanism, because terms can grow exponentially with the number of $β$-steps. This is why implementations of functional languages and proof assistants always rely on some form of sharing of subterms. These frameworks however do not only evaluate $λ$-terms, they also have to compare them for equality. In presence of sharing, one is actually interested in the equality---or more precisely $α$-conversion---of the underlying unshared $λ$-terms. The literature contains algorithms for such a sharing equality, that are polynomial in the sizes of the shared terms. This paper improves the bounds in the literature by presenting the first linear time algorithm. As others before us, we are inspired by Paterson and Wegman's algorithm for first-order unification, itself based on representing terms with sharing as DAGs, and sharing equality as bisimulation of DAGs.

cs.LO↗

arXiv2015

On the Relative Usefulness of Fireballs

In CSL-LICS 2014, Accattoli and Dal Lago showed that there is an implementation of the ordinary (i.e. strong, pure, call-by-name) $λ$-calculus into models like RAM machines which is polynomial in the number of $β$-steps, answering a long-standing question. The key ingredient was the use of a calculus with useful sharing, a new notion whose complexity was shown to be polynomial, but whose implementation was not explored. This paper, meant to be complementary, studies useful sharing in a call-by-value scenario and from a practical point of view. We introduce the Fireball Calculus, a natural extension of call-by-value to open terms for which the problem is as hard as for the ordinary lambda-calculus. We present three results. First, we adapt the solution of Accattoli and Dal Lago, improving the meta-theory of useful sharing. Then, we refine the picture by introducing the GLAMoUr, a simple abstract machine implementing the Fireball Calculus extended with useful sharing. Its key feature is that usefulness of a step is tested---surprisingly---in constant time. Third, we provide a further optimization that leads to an implementation having only a linear overhead with respect to the number of $β$-steps.

cs.LO↗

arXiv2012

On the correctness of a branch displacement algorithm

The branch displacement problem is a well-known problem in assembler design. It revolves around the feature, present in several processor families, of having different instructions, of different sizes, for jumps of different displacements. The problem, which is provably NP-hard, is then to select the instructions such that one ends up with the smallest possible program. During our research with the CerCo project on formally verifying a C compiler, we have implemented and proven correct an algorithm for this problem. In this paper, we discuss the problem, possible solutions, our specific solutions and the proofs.

cs.LO↗

arXiv2012

A Bi-Directional Refinement Algorithm for the Calculus of (Co)Inductive Constructions

The paper describes the refinement algorithm for the Calculus of (Co)Inductive Constructions (CIC) implemented in the interactive theorem prover Matita. The refinement algorithm is in charge of giving a meaning to the terms, types and proof terms directly written by the user or generated by using tactics, decision procedures or general automation. The terms are written in an "external syntax" meant to be user friendly that allows omission of information, untyped binders and a certain liberal use of user defined sub-typing. The refiner modifies the terms to obtain related well typed terms in the internal syntax understood by the kernel of the ITP. In particular, it acts as a type inference algorithm when all the binders are untyped. The proposed algorithm is bi-directional: given a term in external syntax and a type expected for the term, it propagates as much typing information as possible towards the leaves of the term. Traditional mono-directional algorithms, instead, proceed in a bottom-up way by inferring the type of a sub-term and comparing (unifying) it with the type expected by its context only at the end. We propose some novel bi-directional rules for CIC that are particularly effective. Among the benefits of bi-directionality we have better error message reporting and better inference of dependent types. Moreover, thanks to bi-directionality, the coercion system for sub-typing is more effective and type inference generates simpler unification problems that are more likely to be solved by the inherently incomplete higher order unification algorithms implemented. Finally we introduce in the external syntax the notion of vector of placeholders that enables to omit at once an arbitrary number of arguments. Vectors of placeholders allow a trivial implementation of implicit arguments and greatly simplify the implementation of primitive and simple tactics.

cs.LO↗

arXiv2011

A Foundational View on Integration Problems

The integration of reasoning and computation services across system and language boundaries is a challenging problem of computer science. In this paper, we use integration for the scenario where we have two systems that we integrate by moving problems and solutions between them. While this scenario is often approached from an engineering perspective, we take a foundational view. Based on the generic declarative language MMT, we develop a theoretical framework for system integration using theories and partial theory morphisms. Because MMT permits representations of the meta-logical foundations themselves, this includes integration across logics. We discuss safe and unsafe integration schemes and devise a general form of safe integration.

cs.LO↗

arXiv2011

Nonuniform Coercions via Unification Hints

We introduce the notion of nonuniform coercion, which is the promotion of a value of one type to an enriched value of a different type via a nonuniform procedure. Nonuniform coercions are a generalization of the (uniform) coercions known in the literature and they arise naturally when formalizing mathematics in an higher order interactive theorem prover using convenient devices like canonical structures, type classes or unification hints. We also show how nonuniform coercions can be naturally implemented at the user level in an interactive theorem prover that allows unification hints.

cs.LO↗

arXiv2010

General Recursion and Formal Topology

It is well known that general recursion cannot be expressed within Martin-Loef's type theory and various approaches have been proposed to overcome this problem still maintaining the termination of the computation of the typable terms. In this work we propose a new approach to this problem based on the use of inductively generated formal topologies.

cs.LO↗