Skip to content
Conference Open access

Enrich, Aggregate, and Generate: Three-stage Biomedical Data-to-Text Generation Using Large Language Models in Low-resource Scenarios

2026 · Annual Meeting of the Association for Computational Linguistics · pp. 37611-37622 · 0 citations · 46 references
Computer Science

TL;DR

This paper proposes an EAG (En-rich, Aggregate, and Generate) framework, a simple but efficient LLM-based three-stage biomedical D2T approach in low-resource scenarios, and conducts extensive evaluations of closed-source general LLMs, open-source general LLMs, and open-source medical LLMs.

Abstract

Biomedical data-to-text generation aims at generating textual natural language descriptions that can fluently and precisely describe the biomedical structured data. However, biomedical data-to-text generation faces the dilemma of a lack of labeled data due to the privacy and scarcity of medical data. Large language models (LLMs) have demonstrated the ability to solve few-shot tasks through in-context learning (ICL). In this paper, we are the first to explore the performance of different LLMs in the biomedical data-to-text generation task. To address the issues of semantic sparsity and mis-interpretation of numerical values in biomedical structured data, we propose an EAG (En-rich, Aggregate, and Generate) framework, a simple but efficient LLM-based three-stage biomedical D2T approach in low-resource scenarios. We conduct extensive evaluations of closed-source general LLMs, open-source general LLMs, and open-source medical LLMs. The results show that EAG framework provides good interpretability and superior performance, achieving state-of-the-art performance on the BioLeaflets dataset. The code and data will be released at https://github.com/FXLP/EAG.

Read PDF

Similar papers

Preprint Jul 2026

Exploring Post-Training Alignment of Small Language Models for Biomedical Data-to-Text Generation: A Case Study of Medication Leaflet

Translating complex biomedical data into patient-friendly narratives is central to modern biomedical informatics. This study presents a comparative analysis of training small language models (SLMs) in specialized biomedical datato-text generation tasks. We explore widely adopted post-training methods including supervised fine-tuning (SFT), direct preference optimization (DPO), odds ratio preference optimization (ORPO), and group relative policy optimization (GRPO) with Qwen-based SLMs on a medicine package leaflets dataset. To assess cross-dataset generalizability, we also curated drug label data from openFDA. We evaluate models using both standard lexical overlap metrics like ROUGE as well as semantic similarity measures. Across our experiments, the results show that (1) the aligned SLMs outperform proprietary models like GPT-5; (2) ORPO outperforms the SFTbaselines; (3) GRPO yields the most robust cross-dataset performance among the alignment methods tested as well as GPT-5.

Xi Yang, Guodong Liu, Chuqin Li et al. · 0 citations
#small language model Preprint Aug 2026

Cross-Domain, Multi-Task Data-to-Text Generation without In-Domain Training Data

This work compares data-driven knowledge distillation (DDKD) against zero-shot inference and fine-tuning on out-of-domain D2T data, and introduces structure-preserving augmentation via structural subsampling and perturbation in cross-domain D2T generation.

Yifei Song, Kun Efimov-Zhang, Claire Gardent · 0 citations
Review Open access Aug 2026

NATURAL LANGUAGE PROCESSING USING DEEP LEARNING: A COMPREHENSIVE REVIEW OF MODELS, TECHNIQUES, AND FUTURE DIRECTIONS

Natural language processing (NLP) has emerged as a key focus of AI research for the analysis, interpretation, extraction, summarisation, and generation of human language. The vast amount of unstructured textual data in scientific research, electronic health records, clinical notes, radiology reports, public health documents, and digital health platforms has driven the demand for sophisticated computational tools and techniques capable of extracting structured and actionable knowledge from language. NLP has been greatly advanced by deep learning, which allows for automatic representation learning, understanding context, modeling sequences, and generating large amounts of language by means of structures like CNN, RNN, LSTM, GRU, attention mechanisms, transformers, and large language models. This review aims to present a detailed overview of deep learning-based NLP models, methods, applications, challenges, and future directions, focusing on biomedical informatics, clinical text mining, digital health and biomathematical relevance. It has numerous applications such as biomedical literature mining, named entity recognition, relation extraction, clinical decision support, pharmacovigilance, radiology report generation, public health surveillance, and construction of knowledge graph. The specific focus lies in the application of NLP to identify biological entities, clinical variables and quantitative evidence that can be used to support biomathematical modeling. There are several current challenges such as domain shift, privacy, hallucination, bias, interpretability, and reproducibility. The success of future progress relies on reliable, comprehensible, domain specific and clinically verified NLP systems.

Dr. Pradeep Kumar Atulker, Dr. Rahul Kumar Hindustani, Ravi Shankar Nanduri et al. · 0 citations
Conference Jul 2026

Research on abstract generation methods of large language models in low-resource scenarios optimized based on computer technology

This paper focuses on the text generation task in the field of computer natural language processing, and studies the fast recognition method of large language models in low-resource scenarios. It mainly addresses the low-resource problems such as incomplete annotated data, domain mismatch, and weak model generalization. Firstly, it analyzes the problem of abstract enhancement in low-resource scenarios, and compares the computational adaptability between traditional extractive or generative methods and large language models. Secondly, four key technical schemes oriented to computer model optimization are proposed, including prompt engineering enhancement, parameter-efficient fine-tuning, data augmentation, and semi-supervised pseudo-labeling. Finally, experiments are carried out on scientific and technological, legal, and medical datasets under three low-resource settings: few-shot, zero-shot, and cross-domain. Experimental results show that the proposed computer technology optimization strategy can improve the model convergence speed and generalization ability, with the ROUGE-L index increased by up to 12.6 percentage points, indicating that the technical improvement is effective. This paper can provide a reference for the computer technology optimization of natural language processing tasks in low-resource scenarios.

Yanji Li · 0 citations
Review Open access Jul 2026

Domain-specific versus general large language models: a review and empirical benchmark in real medical texts

It is suggested that domain-adapted encoder models may be preferable for similar structured clinical NER settings, although larger and externally validated benchmarks are needed before generalizing to other languages, clinical corpora, model families, or deployment environments.

L. Elvas, Carolina Carvalho · 0 citations