Although AI systems are becoming increasingly common in the workplace, research on their integration into human teams remains limited. In particular, little is known about how the embodiment of artificial agents shapes collaboration and performance in non-routine analytical tasks. To address this gap, we examine how different degrees of embodiment affect team performance and conversational dynamics in a real-life escape room. Teams composed of either three humans or two humans and an artificial agent (a Box, an Avatar, or a hyper-realistic humanoid) worked together to escape the room within a time limit. Our findings show that artificial agents have an uneven impact on team outcomes, with some mixed human–AI teams performing exceptionally well and others markedly worse. Human-only teams, by contrast, display more consistent performance: they are more likely to complete all tasks successfully, although they take longer and commit more errors. We also document a suggestive non-linear relationship between embodiment and team performance. Teams interacting with more embodied agents display conversational patterns that more closely resemble human–human dialogue. Together, these findings show that embodied AI shapes collaboration in complex ways, reinforcing evidence that social cues critically guide teamwork dynamics.

In this paper, we explore the impact of augmenting pre-trained Encoder-Decoder models, specifically T5, with linguistic knowledge for the prediction of a target task. In particular, we investigate whether fine-tuning a T5 model on an intermediate task that predicts structural linguistic properties of sentences modifies its performance in the target task of predicting sentence-level complexity. Our study encompasses diverse experiments conducted on Italian and English datasets, employing both monolingual and multilingual T5 models at various sizes. Results obtained for both languages and in cross-lingual configurations show that linguistically motivated intermediate fine-tuning has generally a positive impact on target task performance, especially when applied to smaller models and in scenarios with limited data availability.

Large Language Models (LLMs) undergo extensive evaluation against various benchmarks collected in established leaderboards to assess their performance across multiple tasks.However, to the best of our knowledge, there is a lack of comprehensive studies evaluating these models' linguistic abilities independent of specific tasks.In this paper, we introduce a novel evaluation methodology designed to test LLMs' sentence generation abilities under specific linguistic constraints.Drawing on the 'linguistic profiling' approach, we rigorously investigate the extent to which five LLMs of varying sizes, tested in both zero-and few-shot scenarios, effectively adhere to (morpho)syntactic constraints.Our findings shed light on the linguistic proficiency of LLMs, revealing both their capabilities and limitations in generating linguistically-constrained sentences 1 .

In this paper, we propose a comprehensive linguistic study aimed at assessing the implicit behavior of one of the most prominent Neural Language Models (NLM) based on Transformer architectures, BERT Devlin et al., when dealing with a particular source of noisy data, namely essays written by L1 Italian learners containing a variety of errors targeting grammar, orthography and lexicon. Differently from previous works, we focus on the pre-training stage and we devise two complementary evaluation tasks aimed at assessing the impact of errors on sentence-level inner representations in terms of semantic robustness and linguistic sensitivity. While the first evaluation perspective is meant to probe the model's ability to encode the semantic similarity between sentences also in the presence of errors, the second type of probing task evaluates the influence of errors on BERT's implicit knowledge of a set of raw and morpho-syntactic properties of a sentence. Our experiments show that BERT's ability to compute sentence similarity and to correctly encode multi-leveled linguistic information of a sentence are differently modulated by the category of errors and that the error hierarchies in terms of robustness and sensitivity change across layer-wise representations.

Several studies in the literature on the interpretation of Neural Language Models (NLM) focus on the linguistic generalization abilities of pre-trained models. However, little attention is paid to how the linguistic knowledge of the models changes during the fine-tuning steps. In this paper, we contribute to this line of research by showing to what extent a wide range of linguistic phenomena are forgotten across 50 epochs of fine-tuning, and how the preserved linguistic knowledge is correlated with the resolution of the fine-tuning task. To this end, we considered a quite understudied task where linguistic information plays the main role, i.e. the prediction of the evolution of written language competence of native language learners. In addition, we investigate whether it is possible to predict the fine-tuned NLM accuracy across the 50 epochs solely relying on the assessed linguistic competence. Our results are encouraging and show a high relationship between the model’s linguistic competence and its ability to solve a linguistically-based downstream task.

In this paper we investigate the linguistic knowledge learned by a Neural Language Model (NLM) before and after a fine-tuning process and how this knowledge affects its predictions during several classification problems.We use a wide set of probing tasks, each of which corresponds to a distinct sentence-level feature extracted from different levels of linguistic annotation.We show that BERT is able to encode a wide range of linguistic characteristics, but it tends to lose this information when trained on specific downstream tasks.We also find that BERT's capacity to encode different kind of linguistic properties has a positive influence on its predictions: the more it stores readable linguistic information of a sentence, the higher will be its capacity of predicting the expected label assigned to that sentence.

“Voices of the Great War” is the first large corpus of Italian historical texts dating back to the period of First World War. This corpus differs from other existing resources in several respects. First, from the linguistic point of view it gives account of the wide range of varieties in which Italian was articulated in that period, namely from a diastratic (educated vs. uneducated writers), diaphasic (low/informal vs. high/formal registers) and diatopic (regional varieties, dialects) points of view. From the historical perspective, through a collection of texts belonging to different genres it represents different views on the war and the various styles of narrating war events and experiences. The final corpus is balanced along various dimensions, corresponding to the textual genre, the language variety used, the author type and the typology of conveyed contents. The corpus is fully annotated with lemmas, part-of-speech, terminology, and named entities. Significant corpus samples representative of the different “voices” have also been enriched with meta-linguistic and syntactic information. The layer of syntactic annotation forms the first nucleus of an Italian historical treebank complying with the Universal Dependencies standard. The paper illustrates the final resource, the methodology and tools used to build it, and the Web Interface for navigating it.

In this paper, we introduce Profiling–UD, a new text analysis tool inspired to the principles of linguistic profiling that can support language variation research from different perspectives. It allows the extraction of more than 130 features, spanning across different levels of linguistic description. Beyond the large number of features that can be monitored, a main novelty of Profiling–UD is that it has been specifically devised to be multilingual since it is based on the Universal Dependencies framework. In the second part of the paper, we demonstrate the effectiveness of these features in a number of theoretical and applicative studies in which they were successfully used for text and author profiling.

In this paper, we present a crowdsourcing-based approach to model the human perception of sentence complexity. We collect a large corpus of sentences rated with judgments of complexity for two typologically-different languages, Italian and English. We test our approach in two experimental scenarios aimed to investigate the contribution of a wide set of lexical, morpho-syntactic and syntactic phenomena in predicting i) the degree of agreement among annotators independently from the assigned judgment and ii) the perception of sentence complexity.

In this paper we present PaCCSS-IT, a Parallel Corpus of Complex-Simple Sentences for ITalian.To build the resource we develop a new method for automatically acquiring a corpus of complex-simple paired sentences able to intercept structural transformations and particularly suitable for text simplification.The method requires a wide amount of texts that can be easily extracted from the web making it suitable also for less-resourced languages.We test it on the Italian language making available the biggest Italian corpus for automatic text simplification.

BACKGROUND: Due to the rapidly expanding body of biomedical literature, biologists require increasingly sophisticated and efficient systems to help them to search for relevant information. Such systems should account for the multiple written variants used to represent biomedical concepts, and allow the user to search for specific pieces of knowledge (or events) involving these concepts, e.g., protein-protein interactions. Such functionality requires access to detailed information about words used in the biomedical literature. Existing databases and ontologies often have a specific focus and are oriented towards human use. Consequently, biological knowledge is dispersed amongst many resources, which often do not attempt to account for the large and frequently changing set of variants that appear in the literature. Additionally, such resources typically do not provide information about how terms relate to each other in texts to describe events. RESULTS: This article provides an overview of the design, construction and evaluation of a large-scale lexical and conceptual resource for the biomedical domain, the BioLexicon. The resource can be exploited by text mining tools at several levels, e.g., part-of-speech tagging, recognition of biomedical entities, and the extraction of events in which they are involved. As such, the BioLexicon must account for real usage of words in biomedical texts. In particular, the BioLexicon gathers together different types of terms from several existing data resources into a single, unified repository, and augments them with new term variants automatically extracted from biomedical literature. Extraction of events is facilitated through the inclusion of biologically pertinent verbs (around which events are typically organized) together with information about typical patterns of grammatical and semantic behaviour, which are acquired from domain-specific texts. In order to foster interoperability, the BioLexicon is modelled using the Lexical Markup Framework, an ISO standard. CONCLUSIONS: The BioLexicon contains over 2.2 M lexical entries and over 1.8 M terminological variants, as well as over 3.3 M semantic relations, including over 2 M synonymy relations. Its exploitation can benefit both application developers and users. We demonstrate some such benefits by describing integration of the resource into a number of different tools, and evaluating improvements in performance that this can bring.

The extraction of information from texts requires resources that contain both syntactic and semantic properties of lexical units. As the use of language in specialized domains, such as biology, can be very different to the general domain, there is a need for domain-specific resources to ensure that the information extracted is as accurate as possible. We are building a large-scale lexical resource for the biology domain, providing information about predicate-argument structure that has been bootstrapped from a biomedical corpus on the subject of E. Coli. The lexicon is currently focussed on verbs, and includes both automatically-extracted syntactic subcategorization frames, as well as semantic event frames that are based on annotation by domain experts. In addition, the lexicon contains manually-added explicit links between semantic and syntactic slots in corresponding frames. To our knowledge, this lexicon currently represents a unique resource within in the biomedical domain.

The paper describes a system for the automatic consolidation of Italian legislative texts to be used as a support of an editorial consolidating activity and dealing with the following typology of textual amendments: repeal, substitution and integration. The focus of the paper is on the semantic analysis of the textual amendment provisions and the formalized representation of the amendments in terms of meta-data. The proposed approach to consolidation is metadata--oriented and based on Natural Language Processing (NLP) techniques: we use XML--based standards for metadata annotation of legislative acts and a flexible NLP architecture for extracting metadata from parsed texts. An evaluation of achieved results is also provided.