arXiv subjects

Andrea Zugarini

Publications and source records attributed to Andrea Zugarini.

At least 19 recordsLinked to original sources

arXiv2026

Multilingual Training and Evaluation Resources for Vision-Language Models

Vision Language Models (VLMs) achieved rapid progress in the recent years. However, despite their growth, VLMs development is heavily grounded on English, leading to two main limitations: (i) the lack of multilingual and multimodal datasets for training, and (ii) the scarcity of comprehensive evaluation benchmarks across languages. In this work, we address these gaps by introducing a new comprehensive suite of resources for VLMs training and evaluation spanning five European languages (English, French, German, Italian, and Spanish). We adopt a regeneration-translation paradigm that produces high-quality cross-lingual resources by combining curated synthetic generation and manual annotation. Specifically, we build Multi-PixMo, a training corpus obtained regenerating examples from Pixmo pre-existing datasets with permissively licensed models: PixMo-Cap, PixMo-AskModelAnything, and CoSyn-400k. On the evaluation side, we construct a set of multilingual benchmarks derived translating widely used English datasets (MMbench, ScienceQA, MME, POPE, AI2D). We assess the quality of these resources through qualitative and quantitative human analyses, measuring inter-annotator agreement. Additionally, we perform ablation studies to demonstrate the impact of multilingual data, with respect to English only, in VLMs training. Experiments, comprising 3 different models show that using multilingual, multimodal examples for training VLMs aids is consistently beneficial on non-English benchmarks, with positive transfer to English as well.

cs.CL↗

arXiv2025

Challenging the Abilities of Large Language Models in Italian: a Community Initiative

The rapid progress of Large Language Models (LLMs) has transformed natural language processing and broadened its impact across research and society. Yet, systematic evaluation of these models, especially for languages beyond English, remains limited. "Challenging the Abilities of LAnguage Models in ITAlian" (CALAMITA) is a large-scale collaborative benchmarking initiative for Italian, coordinated under the Italian Association for Computational Linguistics. Unlike existing efforts that focus on leaderboards, CALAMITA foregrounds methodology: it federates more than 80 contributors from academia, industry, and the public sector to design, document, and evaluate a diverse collection of tasks, covering linguistic competence, commonsense reasoning, factual consistency, fairness, summarization, translation, and code generation. Through this process, we not only assembled a benchmark of over 20 tasks and almost 100 subtasks, but also established a centralized evaluation pipeline that supports heterogeneous datasets and metrics. We report results for four open-weight LLMs, highlighting systematic strengths and weaknesses across abilities, as well as challenges in task-specific evaluation. Beyond quantitative results, CALAMITA exposes methodological lessons: the necessity of fine-grained, task-representative metrics, the importance of harmonized pipelines, and the benefits and limitations of broad community engagement. CALAMITA is conceived as a rolling benchmark, enabling continuous integration of new tasks and models. This makes it both a resource -- the most comprehensive and diverse benchmark for Italian to date -- and a framework for sustainable, community-driven evaluation. We argue that this combination offers a blueprint for other languages and communities seeking inclusive and rigorous LLM evaluation practices.

cs.CL↗

arXiv2025

Protoknowledge Shapes Behaviour of LLMs in Downstream Tasks: Memorization and Generalization with Knowledge Graphs

We introduce the concept of protoknowledge to formalize and measure how sequences of tokens encoding Knowledge Graphs are internalized during pretraining and utilized at inference time by Large Language Models (LLMs). Indeed, LLMs have demonstrated the ability to memorize vast amounts of token sequences during pretraining, and a central open question is how they leverage this memorization as reusable knowledge through generalization. We then categorize protoknowledge into lexical, hierarchical, and topological forms, varying on the type of knowledge that needs to be activated. We measure protoknowledge through Knowledge Activation Tasks (KATs), analyzing its general properties such as semantic bias. We then investigate the impact of protoknowledge on Text-to-SPARQL performance by varying prompting strategies depending on input conditions. To this end, we adopt a novel analysis framework that assesses whether model predictions align with the successful activation of the relevant protoknowledge for each query. This methodology provides a practical tool to explore Semantic-Level Data Contamination and serves as an effective strategy for Closed-Pretraining models.

cs.CL↗

arXiv2025

MSTS: A Multimodal Safety Test Suite for Vision-Language Models

Vision-language models (VLMs), which process image and text inputs, are increasingly integrated into chat assistants and other consumer AI applications. Without proper safeguards, however, VLMs may give harmful advice (e.g. how to self-harm) or encourage unsafe behaviours (e.g. to consume drugs). Despite these clear hazards, little work so far has evaluated VLM safety and the novel risks created by multimodal inputs. To address this gap, we introduce MSTS, a Multimodal Safety Test Suite for VLMs. MSTS comprises 400 test prompts across 40 fine-grained hazard categories. Each test prompt consists of a text and an image that only in combination reveal their full unsafe meaning. With MSTS, we find clear safety issues in several open VLMs. We also find some VLMs to be safe by accident, meaning that they are safe because they fail to understand even simple test prompts. We translate MSTS into ten languages, showing non-English prompts to increase the rate of unsafe model responses. We also show models to be safer when tested with text only rather than multimodal prompts. Finally, we explore the automation of VLM safety assessments, finding even the best safety classifiers to be lacking.

cs.CL↗

arXiv2024

SLIMER-IT: Zero-Shot NER on Italian Language

Traditional approaches to Named Entity Recognition (NER) frame the task into a BIO sequence labeling problem. Although these systems often excel in the downstream task at hand, they require extensive annotated data and struggle to generalize to out-of-distribution input domains and unseen entity types. On the contrary, Large Language Models (LLMs) have demonstrated strong zero-shot capabilities. While several works address Zero-Shot NER in English, little has been done in other languages. In this paper, we define an evaluation framework for Zero-Shot NER, applying it to the Italian language. Furthermore, we introduce SLIMER-IT, the Italian version of SLIMER, an instruction-tuning approach for zero-shot NER leveraging prompts enriched with definition and guidelines. Comparisons with other state-of-the-art models, demonstrate the superiority of SLIMER-IT on never-seen-before entity tags.

cs.CL↗

arXiv2024

Show Less, Instruct More: Enriching Prompts with Definitions and Guidelines for Zero-Shot NER

Recently, several specialized instruction-tuned Large Language Models (LLMs) for Named Entity Recognition (NER) have emerged. Compared to traditional NER approaches, these models have demonstrated strong generalization capabilities. Existing LLMs primarily focus on addressing zero-shot NER on Out-of-Domain inputs, while fine-tuning on an extensive number of entity classes that often highly or completely overlap with test sets. In this work instead, we propose SLIMER, an approach designed to tackle never-seen-before entity tags by instructing the model on fewer examples, and by leveraging a prompt enriched with definition and guidelines. Experiments demonstrate that definition and guidelines yield better performance, faster and more robust learning, particularly when labelling unseen named entities. Furthermore, SLIMER performs comparably to state-of-the-art approaches in out-of-domain zero-shot NER, while being trained in a more fair, though certainly more challenging, setting.

cs.CL↗

arXiv2024

Caption-Driven Explorations: Aligning Image and Text Embeddings through Human-Inspired Foveated Vision

Understanding human attention is crucial for vision science and AI. While many models exist for free-viewing, less is known about task-driven image exploration. To address this, we introduce CapMIT1003, a dataset with captions and click-contingent image explorations, to study human attention during the captioning task. We also present NevaClip, a zero-shot method for predicting visual scanpaths by combining CLIP models with NeVA algorithms. NevaClip generates fixations to align the representations of foveated visual stimuli and captions. The simulated scanpaths outperform existing human attention models in plausibility for captioning and free-viewing tasks. This research enhances the understanding of human attention and advances scanpath prediction models.

cs.CV↗

arXiv2024

Dynamic Few-Shot Learning for Knowledge Graph Question Answering

Large language models present opportunities for innovative Question Answering over Knowledge Graphs (KGQA). However, they are not inherently designed for query generation. To bridge this gap, solutions have been proposed that rely on fine-tuning or ad-hoc architectures, achieving good results but limited out-of-domain distribution generalization. In this study, we introduce a novel approach called Dynamic Few-Shot Learning (DFSL). DFSL integrates the efficiency of in-context learning and semantic similarity and provides a generally applicable solution for KGQA with state-of-the-art performance. We run an extensive evaluation across multiple benchmark datasets and architecture configurations.

cs.CL↗

arXiv2024

Clue-Instruct: Text-Based Clue Generation for Educational Crossword Puzzles

Crossword puzzles are popular linguistic games often used as tools to engage students in learning. Educational crosswords are characterized by less cryptic and more factual clues that distinguish them from traditional crossword puzzles. Despite there exist several publicly available clue-answer pair databases for traditional crosswords, educational clue-answer pairs datasets are missing. In this article, we propose a methodology to build educational clue generation datasets that can be used to instruct Large Language Models (LLMs). By gathering from Wikipedia pages informative content associated with relevant keywords, we use Large Language Models to automatically generate pedagogical clues related to the given input keyword and its context. With such an approach, we created clue-instruct, a dataset containing 44,075 unique examples with text-keyword pairs associated with three distinct crossword clues. We used clue-instruct to instruct different LLMs to generate educational clues from a given input content and keyword. Both human and automatic evaluations confirmed the quality of the generated clues, thus validating the effectiveness of our approach.

cs.CL↗

arXiv2024

Multi-word Tokenization for Sequence Compression

Large Language Models have proven highly successful at modelling a variety of tasks. However, this comes at a steep computational cost that hinders wider industrial uptake. In this paper, we present MWT: a Multi-Word Tokenizer that goes beyond word boundaries by representing frequent multi-word expressions as single tokens. MWTs produce a more compact and efficient tokenization that yields two benefits: (1) Increase in performance due to a greater coverage of input data given a fixed sequence length budget; (2) Faster and lighter inference due to the ability to reduce the sequence length with negligible drops in performance. Our results show that MWT is more robust across shorter sequence lengths, thus allowing for major speedups via early sequence truncation.

cs.CL↗

arXiv2024

Are Compressed Language Models Less Subgroup Robust?

To reduce the inference cost of large language models, model compression is increasingly used to create smaller scalable models. However, little is known about their robustness to minority subgroups defined by the labels and attributes of a dataset. In this paper, we investigate the effects of 18 different compression methods and settings on the subgroup robustness of BERT language models. We show that worst-group performance does not depend on model size alone, but also on the compression method used. Additionally, we find that model compression does not always worsen the performance on minority subgroups. Altogether, our analysis serves to further research into the subgroup robustness of model compression.

cs.LG↗

arXiv2024

Neural paraphrasing by automatically crawled and aligned sentence pairs

Paraphrasing is the task of re-writing an input text using other words, without altering the meaning of the original content. Conversational systems can exploit automatic paraphrasing to make the conversation more natural, e.g., talking about a certain topic using different paraphrases in different time instants. Recently, the task of automatically generating paraphrases has been approached in the context of Natural Language Generation (NLG). While many existing systems simply consist in rule-based models, the recent success of the Deep Neural Networks in several NLG tasks naturally suggests the possibility of exploiting such networks for generating paraphrases. However, the main obstacle toward neural-network-based paraphrasing is the lack of large datasets with aligned pairs of sentences and paraphrases, that are needed to efficiently train the neural models. In this paper we present a method for the automatic generation of large aligned corpora, that is based on the assumption that news and blog websites talk about the same events using different narrative styles. We propose a similarity search procedure with linguistic constraints that, given a reference sentence, is able to locate the most similar candidate paraphrases out from millions of indexed sentences. The data generation process is evaluated in the case of the Italian language, performing experiments using pointer-based deep neural architectures.

cs.CL↗

arXiv2024

Fast Vocabulary Transfer for Language Model Compression

Real-world business applications require a trade-off between language model performance and size. We propose a new method for model compression that relies on vocabulary transfer. We evaluate the method on various vertical domains and downstream tasks. Our results indicate that vocabulary transfer can be effectively used in combination with other compression techniques, yielding a significant reduction in model size and inference time while marginally compromising on performance.

cs.CL↗

arXiv2024

BUSTER: a "BUSiness Transaction Entity Recognition" dataset

Albeit Natural Language Processing has seen major breakthroughs in the last few years, transferring such advances into real-world business cases can be challenging. One of the reasons resides in the displacement between popular benchmarks and actual data. Lack of supervision, unbalanced classes, noisy data and long documents often affect real problems in vertical domains such as finance, law and health. To support industry-oriented research, we present BUSTER, a BUSiness Transaction Entity Recognition dataset. The dataset consists of 3779 manually annotated documents on financial transactions. We establish several baselines exploiting both general-purpose and domain-specific language models. The best performing model is also used to automatically annotate 6196 documents, which we release as an additional silver corpus to BUSTER.

cs.CL↗

arXiv2024

An energy-based comparative analysis of common approaches to text classification in the Legal domain

Most Machine Learning research evaluates the best solutions in terms of performance. However, in the race for the best performing model, many important aspects are often overlooked when, on the contrary, they should be carefully considered. In fact, sometimes the gaps in performance between different approaches are neglectable, whereas factors such as production costs, energy consumption, and carbon footprint must take into consideration. Large Language Models (LLMs) are extensively adopted to address NLP problems in academia and industry. In this work, we present a detailed quantitative comparison of LLM and traditional approaches (e.g. SVM) on the LexGLUE benchmark, which takes into account both performance (standard indices) and alternative metrics such as timing, power consumption and cost, in a word: the carbon-footprint. In our analysis, we considered the prototyping phase (model selection by training-validation-test iterations) and in-production phases separately, since they follow different implementation procedures and also require different resources. The results indicate that very often, the simplest algorithms achieve performance very close to that of large LLMs but with very low power consumption and lower resource demands. The results obtained could suggest companies to include additional evaluations in the choice of Machine Learning (ML) solutions.

cs.CL↗

arXiv2023

The WebCrow French Crossword Solver

Crossword puzzles are one of the most popular word games, played in different languages all across the world, where riddle style can vary significantly from one country to another. Automated crossword resolution is challenging, and typical solvers rely on large databases of previously solved crosswords. In this work, we extend WebCrow 2.0, an automatic crossword solver, to French, making it the first program for crossword solving in the French language. To cope with the lack of a large repository of clue-answer crossword data, WebCrow 2.0 exploits multiple modules, called experts, that retrieve candidate answers from heterogeneous resources, such as the web, knowledge graphs, and linguistic rules. We compared WebCrow's performance against humans in two different challenges. Despite the limited amount of past crosswords, French WebCrow was competitive, actually outperforming humans in terms of speed and accuracy, thus proving its capabilities to generalize to new languages.

cs.CL↗

arXiv2023

Contrastive Language-Image Pretrained Models are Zero-Shot Human Scanpath Predictors

Understanding the mechanisms underlying human attention is a fundamental challenge for both vision science and artificial intelligence. While numerous computational models of free-viewing have been proposed, less is known about the mechanisms underlying task-driven image exploration. To address this gap, we present CapMIT1003, a database of captions and click-contingent image explorations collected during captioning tasks. CapMIT1003 is based on the same stimuli from the well-known MIT1003 benchmark, for which eye-tracking data under free-viewing conditions is available, which offers a promising opportunity to concurrently study human attention under both tasks. We make this dataset publicly available to facilitate future research in this field. In addition, we introduce NevaClip, a novel zero-shot method for predicting visual scanpaths that combines contrastive language-image pretrained (CLIP) models with biologically-inspired neural visual attention (NeVA) algorithms. NevaClip simulates human scanpaths by aligning the representation of the foveated visual stimulus and the representation of the associated caption, employing gradient-driven visual exploration to generate scanpaths. Our experimental results demonstrate that NevaClip outperforms existing unsupervised computational models of human visual attention in terms of scanpath plausibility, for both captioning and free-viewing tasks. Furthermore, we show that conditioning NevaClip with incorrect or misleading captions leads to random behavior, highlighting the significant impact of caption guidance in the decision-making process. These findings contribute to a better understanding of mechanisms that guide human attention and pave the way for more sophisticated computational approaches to scanpath prediction that can integrate direct top-down guidance of downstream tasks.

cs.CV↗

arXiv2021

Generate and Revise: Reinforcement Learning in Neural Poetry

Writers, poets, singers usually do not create their compositions in just one breath. Text is revisited, adjusted, modified, rephrased, even multiple times, in order to better convey meanings, emotions and feelings that the author wants to express. Amongst the noble written arts, Poetry is probably the one that needs to be elaborated the most, since the composition has to formally respect predefined meter and rhyming schemes. In this paper, we propose a framework to generate poems that are repeatedly revisited and corrected, as humans do, in order to improve their overall quality. We frame the problem of revising poems in the context of Reinforcement Learning and, in particular, using Proximal Policy Optimization. Our model generates poems from scratch and it learns to progressively adjust the generated text in order to match a target criterion. We evaluate this approach in the case of matching a rhyming scheme, without having any information on which words are responsible of creating rhymes and on how to coherently alter the poem words. The proposed framework is general and, with an appropriate reward shaping, it can be applied to other text generation problems.

cs.CL↗