Which persuasion strategies, if any, are associated with donation compliance? Answering this requires fine-grained strategy labels across a full corpus and statistical tests corrected for multiple comparisons. We annotate all 10,600 persuader turns in the 1,017-dialogue PersuasionForGood corpus with a taxonomy of 41 strategies in 11 categories, using three open-source large language models (Qwen3:30b, Mistral-Small-3.2, Phi-4). Strategy categories alone explain little variance in donation outcome (pseudo R-squared approximately 0.015, consistent across all three annotators). Guilt Induction is the only strategy significantly associated with lower donation rates (approximately -23 percentage points), an effect that replicates across all three models despite only moderate inter-model agreement. Reciprocity is the most robust positive correlate. Target sentiment and interest predict whether a donation occurs but show at most a weak correlation with donation amount. Logistic regression with sentiment, interest, Guilt Induction, and Reciprocity achieves nearly the same fit (pseudo R-squared = 0.080) as the full model with all strategy categories. These findings suggest that strategy identification alone is insufficient to explain persuasion effectiveness, and that guilt-based appeals may be counterproductive in prosocial settings. We release the fully annotated corpus as a public resource.

Grammar refers to the system of rules that governs the structural organization and the semantic relations among linguistic units such as sentences, phrases, and words within a given language. In natural language processing, there remains a notable scarcity of grammar-focused evaluation protocols, a gap that is even more pronounced for low-resource languages. Moreover, the extent to which large language models genuinely comprehend grammatical structure, especially the mapping between syntactic structures and meanings remains under debate. To investigate this issue, we propose a Grammar-Book–Guided evaluation pipeline intended to provide a systematic and generalizable framework for grammar evaluation consisting of four key stages, and in this work we take Luxembourgish as a case study. The results show a weak positive correlation between translation performance and grammatical understanding, indicating that strong translations do not necessarily imply deep grammatical competence. Larger models perform well overall due to their semantic strength but remain weak in morphology and syntax, struggling particularly with Minimal Pair tasks, while strong reasoning ability offers a promising way to enhance their grammatical understanding.

Small language models (SLMs) offer computationally efficient alternatives to large language models, yet their translation quality for low-resource languages (LRLs) remains severely limited. This work presents the first large-scale evaluation of SLMs across 200 languages, revealing systematic underperformance in LRLs and identifying key sources of linguistic disparity. We show that knowledge distillation from strong teacher models using predominantly monolingual LRL data substantially boosts SLM translation quality—often enabling 2B–3B models to match or surpass systems up to 70B parameters. Our study highlights three core findings: (1) a comprehensive benchmark exposing the limitations of SLMs on 200 languages; (2) evidence that LRL-focused distillation improves translation without inducing catastrophic forgetting, with full-parameter fine-tuning and decoder-only teachers outperforming LoRA and encoder–decoder approaches; and (3) consistent cross-lingual gains demonstrating the scalability and robustness of the method. These results establish an effective, low-cost pathway for improving LRL translation and provide practical guidance for deploying SLMs in truly low-resource settings.

Large Language Models excel at NLP tasks but remain prone to hallucinations, limiting trust in real-world applications. We present HalluGuard, a 4B-parameter Small Reasoning Model (SRM) designed as a guardrail for Retrieval-Augmented Generation (RAG) pipelines, which classify document-claim pairs as grounded or hallucinated in closed-book, document-grounded settings and produces evidence-grounded justifications. Our approach combines (i) a domain-agnostic synthetic dataset derived from FineWeb and refined through multi-stage curation and data reformation, (ii) synthetic grounded and hallucinated claims, and (iii) preference-based fine-tuning with Odds Ratio Preference Optimization (ORPO) to distill large-model reasoning into a smaller backbone. On the RAGTruth subset of the LLM-AggreFact benchmark, HalluGuard achieves 84.4% balanced accuracy (BAcc), surpassing specialized models, MiniCheck (7B; 84.0%) and Granite Guardian 3.3 (8B; 82.2%) while using roughly half their parameters. Across the benchmark, it reaches 77.1% BAcc, surpassing larger general-purpose LLMs such as GPT-4o (75.9%). HalluGuard and datasets will be released upon acceptance.

With the emergence of ChatGPT, Transformer models have significantly advanced text classification and related tasks. Decoder-only models such as Llama exhibit strong performance and flexibility, yet they suffer from inefficiency on inference due to token-by-token generation, and their effectiveness in text classification tasks heavily depends on prompt quality. Moreover, their substantial GPU resource requirements often limit widespread adoption. Thus, the question of whether smaller language models are capable of effectively handling text classification tasks emerges as a topic of significant interest. However, the selection of appropriate models and methodologies remains largely underexplored. In this paper, we conduct a comprehensive evaluation of prompt engineering and supervised fine-tuning methods for transformer-based text classification. Specifically, we focus on practical industrial scenarios, including email classification, legal document categorization, and the classification of extremely long academic texts. We examine the strengths and limitations of smaller models, with particular attention to both their performance and their efficiency in Video Random-Access Memory (VRAM) utilization, thereby providing valuable insights for the local deployment and application of compact models in industrial settings.

Since the United Nations defined the Sustainable Development Goals, studies have shown that these goals are interlinked in different ways. The concept of SDG interlinkages refers to the complex network of interactions existing within and between the SDGs themselves. These interactions are referred to as synergies and trade-offs. Synergies represent positive interactions where the progress of one SDG contributes positively to the progress of another. On the other hand, trade-offs are negative interactions where the progress of one SDG has a negative impact on another. However, evaluating such interlinkages is a complex task, not only because of the multidimensional nature of SDGs, but also because it is highly exposed to personal interpretation bias and technical limitations. Recent studies are mainly based on expert judgements, literature reviews, sentiment or data analysis. To remedy these limitations we propose the use of Small Language Models in addition of an advanced Retrieval Augmented Generation to distinguish synergies and trade-offs between SDGs. In order to validate our results, we have drawn on the study carried out by the European Commission’s Joint Research Centre which provides a database of interlinkages labelled according to the presence of synergies or trade-offs.