Primoz Kocbek

This work describes our submission to the CLEF 2025 SimpleText track Task 1, addressing both sentenceand document-level simplification of scientific texts. The methodology centered on using the gpt-4.1, gpt-4.1mini, and gpt-4.1-nano models from OpenAI. Two distinct approaches were compared: a no-context method relying on prompt engineering and a fine-tuned (FT) method across models. The gpt-4.1-mini model with no-context demonstrated robust performance at both levels of simplification, while the fine-tuned models showed mixed results, highlighting the complexities of simplifying text at different granularities, where gpt-4.1-nano-ft performance stands out at document-level simplification in one case.

Multi-modal retrieval-augmented generation (MM-RAG) promises grounded biomedical QA, but it is unclear when to (i) convert figures/tables into text versus (ii) use optical character recognition (OCR)-free visual retrieval that returns page images and leaves interpretation to the generator. We study this trade-off in glycobiology, a visually dense domain. We built a benchmark of 120 multiple-choice questions (MCQs) from 25 papers, stratified by retrieval difficulty (easy text, medium figures/tables, hard cross-evidence). We implemented four augmentations-None, Text RAG, Multi-modal conversion, and late-interaction visual retrieval (ColPali)-using Docling parsing and Qdrant indexing. We evaluated mid-size open-source and frontier proprietary models (e.g., Gemma-3-27B-IT, GPT-4o family). Additional testing used the GPT-5 family and multiple visual retrievers (ColPali/ColQwen/ColFlor). Accuracy with Agresti-Coull 95% confidence intervals (CIs) was computed over 5 runs per configuration. With Gemma-3-27B-IT, Text and Multi-modal augmentation outperformed OCR-free retrieval (0.722-0.740 vs. 0.510 average accuracy). With GPT-4o, Multi-modal achieved 0.808, with Text 0.782 and ColPali 0.745 close behind; within-model differences were small. In follow-on experiments with the GPT-5 family, the best results with ColPali and ColFlor improved by ~2% to 0.828 in both cases. In general, across the GPT-5 family, ColPali, ColQwen, and ColFlor were statistically indistinguishable. GPT-5-nano trailed larger GPT-5 variants by roughly 8-10%. Pipeline choice is capacity-dependent: converting visuals to text lowers the reader burden and is more reliable for mid-size models, whereas OCR-free visual retrieval becomes competitive under frontier models. Among retrievers, ColFlor offers parity with heavier options at a smaller footprint, making it an efficient default when strong generators are available.

This study investigated the application of Large Language Models (LLMs) for simplifying biomedical texts to enhance health literacy. Using a public dataset, which included plain language adaptations of biomedical abstracts, we developed and evaluated several approaches, specifically a baseline approach using a prompt template, a two AI agent approach, and a fine-tuning approach. We selected OpenAI gpt-4o and gpt-4o mini models as baselines for further research. We evaluated our approaches with quantitative metrics, such as Flesch-Kincaid grade level, SMOG Index, SARI, and BERTScore, G-Eval, as well as with qualitative metric, more precisely 5-point Likert scales for simplicity, accuracy, completeness, brevity. Results showed a superior performance of gpt-4o-mini and an underperformance of FT approaches. G-Eval, a LLM based quantitative metric, showed promising results, ranking the approaches similarly as the qualitative metric.

This study investigated the application of Large Language Models (LLMs) for simplifying biomedical texts to enhance health literacy. Using a public dataset, which included plain language adaptations of biomedical abstracts, we developed and evaluated several approaches, specifically a baseline approach using a prompt template, a two AI agent approach, and a fine-tuning approach. We selected OpenAI gpt-4o and gpt-4o mini models as baselines for further research. We evaluated our approaches with quantitative metrics, such as Flesch-Kincaid grade level, SMOG Index, SARI, and BERTScore, G-Eval, as well as with qualitative metric, more precisely 5-point Likert scales for simplicity, accuracy, completeness, brevity. Results showed a superior performance of gpt-4o-mini and an underperformance of FT approaches. G-Eval, a LLM based quantitative metric, showed promising results, ranking the approaches similarly as the qualitative metric.

Primary care professionals struggle to keep up to date with the latest scientific literature critical in guiding evidence-based practice related to their daily work. To help solve the above-mentioned problem, we employed generative artificial intelligence techniques based on large-scale language models to summarize abstracts of scientific papers. Our objective is to investigate the potential of generative artificial intelligence in diminishing the cognitive load experienced by practitioners, thus exploring its ability to alleviate mental effort and burden. The study participants were provided with two use cases related to preventive care and behavior change, simulating a search for new scientific literature. The study included 113 university students from Slovenia and the United States randomized into three distinct study groups. The first group was assigned to the full abstracts. The second group was assigned to the short abstracts generated by AI. The third group had the option to select a full abstract in addition to the AI-generated short summary. Each use case study included ten retrieved abstracts. Our research demonstrates that the use of generative AI for literature review is efficient and effective. The time needed to answer questions related to the content of abstracts was significantly lower in groups two and three compared to the first group using full abstracts. The results, however, also show significantly lower accuracy in extracted knowledge in cases where full abstract was not available. Such a disruptive technology could significantly reduce the time required for healthcare professionals to keep up with the most recent scientific literature; nevertheless, further developments are needed to help them comprehend the knowledge accurately.

The skin is the largest organ in the human body and is colonized by a diverse microbiota that works in harmony to protect the skin. However, when skin damage occurs, the skin microbiota is also disrupted, and pathogens can invade the wound and cause infection. Probiotics or other beneficial microbes and their metabolites are one possible alternative treatment for combating skin pathogens via their antimicrobial effectiveness. The objective of our study was to evaluate the antimicrobial effect of seven multi-strain dietary supplements and eleven single-strain microbes that contain probiotics against 15 clinical wound pathogens using the agar spot assay, co-culturing assay, and agar well diffusion assay. We also conducted genera-specific and species-specific molecular methods to detect the DNA in the dietary supplements and single-strain beneficial microbes. We found that the multi-strain dietary supplements exhibited a statistically significant higher antagonistic effect against the challenge wound pathogens than the single-strain microbes and that lactobacilli-containing dietary supplements and single-strain microbes were significantly more efficient than the selected propionibacteria and bacilli. Differences in results between methods were also observed, possibly due to different mechanisms of action. Individual pathogens were susceptible to different dietary supplements or single-strain microbes. Perhaps an individual approach such as a ‘probiogram’ could be a possibility in the future as a method to find the most efficient targeted probiotic strains, cell-free supernatants, or neutralized cell-free supernatants that have the highest antagonistic effect against individual clinical wound pathogens.

Type 2 diabetes mellitus (T2DM) often results in high morbidity and mortality. In addition, T2DM presents a substantial financial burden for individuals and their families, health systems, and societies. According to studies and reports, globally, the incidence and prevalence of T2DM are increasing rapidly. Several models have been built to predict T2DM onset in the future or detect undiagnosed T2DM in patients. Additional to the performance of such models, their interpretability is crucial for health experts, especially in personalized clinical prediction models. Data collected over 42 months from health check-up examinations and prescribed drugs data repositories of four primary healthcare providers were used in this study. We propose a framework consisting of LogicRegression based feature extraction and Least Absolute Shrinkage and Selection operator based prediction modeling for undiagnosed T2DM prediction. Performance of the models was measured using Area under the ROC curve (AUC) with corresponding confidence intervals. Results show that using LogicRegression based feature extraction resulted in simpler models, which are easier for healthcare experts to interpret, especially in cases with many binary features. Models developed using the proposed framework resulted in an AUC of 0.818 (95% Confidence Interval (CI): 0.812−0.823) that was comparable to more complex models (i.e., models with a larger number of features), where all features were included in prediction model development with the AUC of 0.816 (95% CI: 0.810−0.822). However, the difference in the number of used features was significant. This study proposes a framework for building interpretable models in healthcare that can contribute to higher trust in prediction models from healthcare experts.

Vending machines in health and social care facilities are often the only possible choice for a quick snack for workers and visitors, in many cases providing unhealthy dietary choices. Our study aimed to analyse the variety and nutritional quality of foods available in vending machines placed in social and health care institution in Slovenia. The available snacks were quantitatively assessed, using traffic light profiling. The model used for nutrient profiling was that of the Food Standards Australia New Zealand (FSANZ). Vending machines in 188 institutions were surveyed, resulting in 5625 food-items consisting of 267 unique product labels. Sweet products dominate in vending machines offers (about 70%), while nuts and seeds (8.4%), yoghurts (2.1%), fruits (1.4%) and milk (0.3%) are present in a very small proportion or are not available at all. According to FSANZ, 88.5% of all displayed food items in vending machines can be considered as lower nutritional quality or less healthy products. The authors’ future activities will be focused on ensuring wider availability of healthy dietary choices and on including official guidelines in tender conditions for vending machines in health and social care institutions in Slovenia.