This work investigates whether small-scale LMs can benefit from instruction tuning (IT). We compare conversational and question–answering IT datasets, applied either in a merged or sequential curriculum, using decoder-only models with 100M and 140M parameters. Evaluation spans both fine-tuning (SuperGLUE) and zero-shot (BLiMP, EWoK, WUGs, entity tracking, and psycholinguistic correlation) settings. Results show that IT yields small but consistent gains in fine-tuning scenarios, with sequential curricula outperforming merged data; however, improvements do not consistently transfer to zero-shot tasks, suggesting a trade-off between interaction-focused adaptation and broad linguistic generalization. These results highlight both the potential and the constraints of adapting human-inspired learning strategies to low-resource LMs, and point toward hybrid, curriculum-based approaches for enhancing generalization under ecological training limits.

Large Language Models (LLMs) are increasingly used in tasks requiring interpretive and inferential accuracy. In this paper, we introduce ExpliCa, a new dataset for evaluating LLMs in explicit causal reasoning. ExpliCa uniquely integrates both causal and temporal relations presented in different linguistic orders and explicitly expressed by linguistic connectives. The dataset is enriched with crowdsourced human acceptability ratings. We tested LLMs on ExpliCa through prompting and perplexity-based metrics. We assessed seven commercial and open-source LLMs, revealing that even top models struggle to reach 0.80 accuracy. Interestingly, models tend to confound temporal relations with causal ones, and their performance is also strongly influenced by the linguistic order of the events. Finally, perplexity-based scores and prompting performance are differently affected by model size.

We introduce MAIA (Multimodal AI Assessment), a native-Italian benchmark designed for fine-grained investigation of the reasoning abilities of visual language models on videos. MAIA differs from other available video benchmarks for its design, its reasoning categories, the metric it uses, and the language and culture of the videos. MAIA evaluates Vision Language Models (VLMs) on two aligned tasks: a visual statement verification task, and an open-ended visual question-answering task, both on the same set of video-related questions. It considers twelve reasoning categories that aim to disentangle language and vision relations by highlighting the role of the visual input. Thanks to its carefully taught design, it evaluates VLMs’ consistency and visually grounded natural language comprehension and generation simultaneously through an aggregated metric revealing low results that highlight models’ fragility. Last but not least, the video collection has been carefully selected to reflect the Italian culture, and the language data are produced by native-speakers.Data available at *[GitHub](https://github.com/Caput97/MAIA-Multimodal_AI_Assessment.git).*

This paper investigates how decoder-only instruction-tuned LLMs handle lexical ambiguity. Two distinct methodologies are employed: Eliciting rating scores from the model via prompting and analysing the cosine similarity between pairs of polysemous words in context. Ratings and embeddings are obtained by providing pairs of sentences from Haber and Poesio (2021) to the model. These ratings and cosine similarity scores are compared with each other and with the human similarity judgments in the dataset.Surprisingly, the model scores show only a moderate correlation with the subjects’ similarity judgments and no correlation with the target word embedding similarities. A vector space anisotropy inspection has also been performed, as a potential source of the experimental results. The analysis reveals that the embedding spaces of two out of the three analyzed models exhibit poor anisotropy, while the third model shows relatively moderate anisotropy compared to previous findings for models with similar architecture (Ethayarajh 2019). These findings offer new insights into the relationship between generation quality and vector representations in decoder-only LLMs.

Multimodal metaphorical interpretation of abstract concepts has always been a debated problem in many research fields, including cognitive linguistics and NLP. With the dramatic improvements of Large Language Models (LLMs) and the increasing attention toward multimodal Vision-Language Models (VLMs), there has been pronounced attention on the conceptualization of abstracts. Nevertheless, a systematic scientific investigation is still lacking. This work introduces a framework designed to shed light on the indirect grounding mechanisms that anchor the meaning of abstract concepts to concrete situations (e.g. ability - a person skating), following the idea that abstracts acquire meaning from embodied and situated simulation. We assessed human and LLMs performances by a situation generation task. Moreover, we assess the figurative richness of images depicting concrete scenarios, via a text-to-image retrieval task performed on LAION-400M.

We present a model for the Strict-Small track of the BabyLM Challenge 2024 (Choshen et al. 2024). We introduce a Curriculum Learning approach for training a specialized version of GPT-2 (Radford et al. 2019), that we name ConcreteGPT. We utilize the norms from (Brysbaert et al. 2014) which provide concreteness ratings for 40,000 English lexical items based on human subjects. Using these norms, we assign a concreteness score to each sentence in the training dataset and develop two curriculum strategies that progressively introduce more complex and abstract language patterns in the training data. Compared to the baselines, our best model shows lower performance on zero-shot tasks but demonstrates superior performance in fine-tuning tasks. Notably, our curriculum-trained models exhibit significant improvements over a non-curriculum based training of the same model.