Source author record

Maurizio Gabbrielli

Maurizio Gabbrielli appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

Artificial Intelligence Machine Learning Programming Languages Computer Vision Software Engineering Distributed, Parallel, and Cluster Computing Graphics Human-Computer Interaction Logic in Computer Science

Catalog footprint

What is connected

14works

9topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2026arXiv

BlenderRAG: High-Fidelity 3D Object Generation via Retrieval-Augmented Code Synthesis

Automatic generation of executable Blender code from natural language remains challenging, with state-of-the-art LLMs producing frequent syntactic errors and geometrically inconsistent objects. We present BlenderRAG, a retrieval-augmented generation system that operates on a curated multimodal dataset of 500 expert-validated examples (text, code, image) across 50 object categories. By retrieving semantically similar examples during generation, BlenderRAG improves compilation success rates from 40.8% to 70.0% and semantic normalized alignment from 0.41 to 0.77 (CLIP similarity) across four state-of-the-art LLMs, without requiring fine-tuning or specialized hardware, making it immediately accessible for deployment. The dataset and code will be available at https://github.com/MaxRondelli/BlenderRAG.

preprint2026arXiv

Do not copy and paste! Rewriting strategies for code retrieval

Embedding-based code retrieval often suffers when encoders overfit to surface syntax. Prior work mitigates this by using LLMs to rephrase queries and corpora into a normalized style, but leaves two questions open: how much representational shift helps, and when is the per-query LLM call justified? We study a hierarchy of three rewriting strategies: stylistic rephrasing, NL-enriched PseudoCode, and full Natural-Language transcription, under joint query-corpus (QC, online) and corpus-only (C, offline) augmentation, across six CoIR benchmarks, five encoders, and three rewriters spanning independent model families (Qwen, DeepSeek, Mistral). We are the first to evaluate NL-enriched PseudoCode and snippet-level Natural Language as direct retrieval representations, rather than as transient intermediates. Full NL rewriting with QC yields the largest gains (+0.51 absolute NDCG@10 on CT-Contest for MoSE-18), while corpus-only rewriting degrades retrieval in 56 of 90 configurations, about 62%. We introduce two diagnostics, Delta H, token entropy, and Delta s, embedding cosine, and show that Delta H predicts retrieval gain under QC across all three rewriter families: pooled Spearman rho = +0.436, p < 0.001 on DeepSeek+Codestral; rho = +0.593 on Codestral alone; rho = +0.356 on Qwen. This establishes Delta H as a cheap, rewriter-agnostic proxy for deciding when rewriting pays off before running retrieval. Our analysis reframes LLM rewriting as a cost-benefit decision: it is most effective as a remediation layer for lightweight encoders on code-dominant queries, with diminishing returns for strong encoders or NL-heavy queries.

preprint2026arXiv

Improving Diffusion Posterior Samplers with Lagged Temporal Corrections for Image Restoration

Diffusion-based posterior sampling (PS) is a leading framework for imaging inverse problems, combining learned priors with measurement constraints. Yet, its standard formulations rely on instantaneous data-consistent estimates, which induce temporal variability in the reverse dynamics. We reinterpret PS from a dynamical perspective, showing that the standard PS update corresponds to a first-order discretization of the diffusion dynamics plus a residual correction capturing the mismatch between the denoised prediction and the data-consistent estimate. A second-order discretization, however, naturally introduces a temporal correction based on the variation of consecutive estimates. Building on this, we propose LAMP, combining the second-order update with the residual correction characterizing a PS technique. LAMP thus inherits a lagged temporal correction, and it can be implemented as a modular plug-in over the PS backbone. We show that LAMP preserves the structure of a posterior sampler, and we perform a one-step risk analysis to characterize when LAMP improves the reverse transition via a bias-variance trade-off. Experiments across multiple imaging tasks demonstrate consistent improvements over strong baselines such as DiffPIR and DDRM, without increasing the number of denoising evaluations.

preprint2021arXiv

Content-Based Textual File Type Detection at Scale

Programming language detection is a common need in the analysis of large source code bases. It is supported by a number of existing tools that rely on several features, and most notably file extensions, to determine file types. We consider the problem of accurately detecting the type of files commonly found in software code bases, based solely on textual file content. Doing so is helpful to classify source code that lack file extensions (e.g., code snippets posted on the Web or executable scripts), to avoid misclassifying source code that has been recorded with wrong or uncommon file extensions, and also shed some light on the intrinsic recognizability of source code files. We propose a simple model that (a) use a language-agnostic word tokenizer for textual files, (b) group tokens in 1-/2-grams, (c) build feature vectors based on N-gram frequencies, and (d) use a simple fully connected neural network as classifier. As training set we use textual files extracted from GitHub repositories with at least 1000 stars, using existing file extensions as ground truth. Despite its simplicity the proposed model reaches 85% in our experiments for a relatively high number of recognized classes (more than 130 file types).

preprint2015arXiv

A Multicore Tool for Constraint Solving

*** To appear in IJCAI 2015 proceedings *** In Constraint Programming (CP), a portfolio solver uses a variety of different solvers for solving a given Constraint Satisfaction / Optimization Problem. In this paper we introduce sunny-cp2: the first parallel CP portfolio solver that enables a dynamic, cooperative, and simultaneous execution of its solvers in a multicore setting. It incorporates state-of-the-art solvers, providing also a usable and configurable framework. Empirical results are very promising. sunny-cp2 can even outperform the performance of the oracle solver which always selects the best solver of the portfolio for a given problem.

preprint2015arXiv

Dynamic Choreographies - Safe Runtime Updates of Distributed Applications

Programming distributed applications free from communication deadlocks and races is complex. Preserving these properties when applications are updated at runtime is even harder. We present DIOC, a language for programming distributed applications that are free from deadlocks and races by construction. A DIOC program describes a whole distributed application as a unique entity (choreography). DIOC allows the programmer to specify which parts of the application can be updated. At runtime, these parts may be replaced by new DIOC fragments from outside the application. DIOC programs are compiled, generating code for each site, in a lower-level language called DPOC. We formalise both DIOC and DPOC semantics as labelled transition systems and prove the correctness of the compilation as a trace equivalence result. As corollaries, DPOC applications are free from communication deadlocks and races, even in presence of runtime updates.

preprint2014arXiv

AIOCJ: A Choreographic Framework for Safe Adaptive Distributed Applications

We present AIOCJ, a framework for programming distributed adaptive applications. Applications are programmed using AIOC, a choreographic language suited for expressing patterns of interaction from a global point of view. AIOC allows the programmer to specify which parts of the application can be adapted. Adaptation takes place at runtime by means of rules, which can change during the execution to tackle possibly unforeseen adaptation needs. AIOCJ relies on a solid theory that ensures applications to be deadlock-free by construction also after adaptation. We describe the architecture of AIOCJ, the design of the AIOC language, and an empirical validation of the framework.

preprint2014arXiv

An Empirical Evaluation of Portfolios Approaches for solving CSPs

Recent research in areas such as SAT solving and Integer Linear Programming has shown that the performances of a single arbitrarily efficient solver can be significantly outperformed by a portfolio of possibly slower on-average solvers. We report an empirical evaluation and comparison of portfolio approaches applied to Constraint Satisfaction Problems (CSPs). We compared models developed on top of off-the-shelf machine learning algorithms with respect to approaches used in the SAT field and adapted for CSPs, considering different portfolio sizes and using as evaluation metrics the number of solved problems and the time taken to solve them. Results indicate that the best SAT approaches have top performances also in the CSP field and are slightly more competitive than simple models built on top of classification algorithms.

preprint2014arXiv

An Enhanced Features Extractor for a Portfolio of Constraint Solvers

Recent research has shown that a single arbitrarily efficient solver can be significantly outperformed by a portfolio of possibly slower on-average solvers. The solver selection is usually done by means of (un)supervised learning techniques which exploit features extracted from the problem specification. In this paper we present an useful and flexible framework that is able to extract an extensive set of features from a Constraint (Satisfaction/Optimization) Problem defined in possibly different modeling languages: MiniZinc, FlatZinc or XCSP. We also report some empirical results showing that the performances that can be obtained using these features are effective and competitive with state of the art CSP portfolio techniques.

preprint2014arXiv

SUNNY: a Lazy Portfolio Approach for Constraint Solving

*** To appear in Theory and Practice of Logic Programming (TPLP) *** Within the context of constraint solving, a portfolio approach allows one to exploit the synergy between different solvers in order to create a globally better solver. In this paper we present SUNNY: a simple and flexible algorithm that takes advantage of a portfolio of constraint solvers in order to compute --- without learning an explicit model --- a schedule of them for solving a given Constraint Satisfaction Problem (CSP). Motivated by the performance reached by SUNNY vs. different simulations of other state of the art approaches, we developed sunny-csp, an effective portfolio solver that exploits the underlying SUNNY algorithm in order to solve a given CSP. Empirical tests conducted on exhaustive benchmarks of MiniZinc models show that the actual performance of SUNNY conforms to the predictions. This is encouraging both for improving the power of CSP portfolio solvers and for trying to export them to fields such as Answer Set Programming and Constraint Logic Programming.

preprint2014arXiv

Timed Soft Concurrent Constraint Programs: An Interleaved and a Parallel Approach

We propose a timed and soft extension of Concurrent Constraint Programming. The time extension is based on the hypothesis of bounded asynchrony: the computation takes a bounded period of time and is measured by a discrete global clock. Action prefixing is then considered as the syntactic marker which distinguishes a time instant from the next one. Supported by soft constraints instead of crisp ones, tell and ask agents are now equipped with a preference (or consistency) threshold which is used to determine their success or suspension. In the paper we provide a language to describe the agents behavior, together with its operational and denotational semantics, for which we also prove the compositionality and correctness properties. After presenting a semantics using maximal parallelism of actions, we also describe a version for their interleaving on a single processor (with maximal parallelism for time elapsing). Coordinating agents that need to take decisions both on preference values and time events may benefit from this language. To appear in Theory and Practice of Logic Programming (TPLP).

preprint2013arXiv

Unfolding for CHR programs

Program transformation is an appealing technique which allows to improve run-time efficiency, space-consumption, and more generally to optimize a given program. Essentially, it consists of a sequence of syntactic program manipulations which preserves some kind of semantic equivalence. Unfolding is one of the basic operations which is used by most program transformation systems and which consists in the replacement of a procedure call by its definition. While there is a large body of literature on transformation and unfolding of sequential programs, very few papers have addressed this issue for concurrent languages. This paper defines an unfolding system for CHR programs. We define an unfolding rule, show its correctness and discuss some conditions which can be used to delete an unfolded rule while preserving the program meaning. We also prove that, under some suitable conditions, confluence and termination are preserved by the above transformation. To appear in Theory and Practice of Logic Programming (TPLP)

preprint2011arXiv

On the Expressive Power of Multiple Heads in CHR

Constraint Handling Rules (CHR) is a committed-choice declarative language which has been originally designed for writing constraint solvers and which is nowadays a general purpose language. CHR programs consist of multi-headed guarded rules which allow to rewrite constraints into simpler ones until a solved form is reached. Many empirical evidences suggest that multiple heads augment the expressive power of the language, however no formal result in this direction has been proved, so far. In the first part of this paper we analyze the Turing completeness of CHR with respect to the underneath constraint theory. We prove that if the constraint theory is powerful enough then restricting to single head rules does not affect the Turing completeness of the language. On the other hand, differently from the case of the multi-headed language, the single head CHR language is not Turing powerful when the underlying signature (for the constraint theory) does not contain function symbols. In the second part we prove that, no matter which constraint theory is considered, under some reasonable assumptions it is not possible to encode the CHR language (with multi-headed rules) into a single headed language while preserving the semantics of the programs. We also show that, under some stronger assumptions, considering an increasing number of atoms in the head of a rule augments the expressive power of the language. These results provide a formal proof for the claim that multiple heads augment the expressive power of the CHR language.

preprint2011arXiv

Solving XCSP problems by using Gecode

Gecode is one of the most efficient libraries that can be used for constraint solving. However, using it requires dealing with C++ programming details. On the other hand several formats for representing constraint networks have been proposed. Among them, XCSP has been proposed as a format based on XML which allows us to represent constraints defined either extensionally or intensionally, permits global constraints and has been the standard format of the international competition of constraint satisfaction problems solvers. In this paper we present a plug-in for solving problems specified in XCSP by exploiting the Gecode solver. This is done by dynamically translating constraints into Gecode library calls, thus avoiding the need to interact with C++.

Institution

Affiliation not imported yet

This author record came from a source that does not expose affiliation metadata. Once the author claims the profile or we enrich the record from another provider, this section will link to the concrete institution.

Topic footprint