{"ID":2883135,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.08730","arxiv_id":"2508.08730","title":"Magical: Medical Lay Language Generation via Semantic Invariance and Layperson-tailored Adaptation","abstract":"Medical Lay Language Generation (MLLG) plays a vital role in improving the accessibility of complex scientific content for broader audiences. Recent literature to MLLG commonly employ parameter-efficient fine-tuning methods such as Low-Rank Adaptation (LoRA) to fine-tuning large language models (LLMs) using paired expert-lay language datasets. However, LoRA struggles with the challenges posed by multi-source heterogeneous MLLG datasets. Specifically, through a series of exploratory experiments, we reveal that standard LoRA fail to meet the requirement for semantic fidelity and diverse lay-style generation in MLLG task. To address these limitations, we propose Magical, an asymmetric LoRA architecture tailored for MLLG under heterogeneous data scenarios. Magical employs a shared matrix $A$ for abstractive summarization, along with multiple isolated matrices $B$ for diverse lay-style generation. To preserve semantic fidelity during the lay language generation process, Magical introduces a Semantic Invariance Constraint to mitigate semantic subspace shifts on matrix $A$. Furthermore, to better adapt to diverse lay-style generation, Magical incorporates the Recommendation-guided Switch, an externally interface to prompt the LLM to switch between different matrices $B$. Experimental results on three real-world lay language generation datasets demonstrate that Magical consistently outperforms prompt-based methods, vanilla LoRA, and its recent variants, while also reducing trainable parameters by 31.66%. Our code is publicly available at https://github.com/tianlwang/Magical.git.","short_abstract":"Medical Lay Language Generation (MLLG) plays a vital role in improving the accessibility of complex scientific content for broader audiences. Recent literature to MLLG commonly employ parameter-efficient fine-tuning methods such as Low-Rank Adaptation (LoRA) to fine-tuning large language models (LLMs) using paired expe...","url_abs":"https://arxiv.org/abs/2508.08730","url_pdf":"https://arxiv.org/pdf/2508.08730v2","authors":"[\"Weibin Liao\",\"Tianlong Wang\",\"Yinghao Zhu\",\"Yasha Wang\",\"Junyi Gao\",\"Liantao Ma\"]","published":"2025-08-12T08:21:58Z","proceeding":"cs.CL","tasks":"[\"cs.CL\"]","methods":"[\"Large Language Model\",\"Language Model\",\"LoRA\"]","has_code":false,"code_links":[{"ID":610958,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2883135,"paper_url":"https://arxiv.org/abs/2508.08730","paper_title":"Magical: Medical Lay Language Generation via Semantic Invariance and Layperson-tailored Adaptation","repo_url":"https://github.com/tianlwang/Magical.git","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}