The paper introduces a web‑data curation recipe for pretraining medical encoders, addressing the scarcity of large, diverse corpora in dense‑terminology domains like medicine. It proposes two complementary techniques: medical‑term density filtering to select documents rich in medical terminology, and signal‑amplifying rephrasing that uses an LLM to rewrite documents into denser variants with broader entity contexts. Applied to French medical NLP, the recipe produces the FineMed corpus and the DoctoBERT encoder family, achieving state‑of‑the‑art results on the DrBenchmark public benchmark and a proprietary clinical NER task.
arXiv:2606. 01904v1 Announce Type: cross Abstract: The increasing application of Natural Language Processing (NLP) in healthcare demands language models specifically attuned to the complexities of clinical language.
arXiv:2607. 18927v1 Announce Type: new Abstract: We present OntoBook, a method that converts medical ontology structure into pretraining signal for encoder language models.
arXiv:2608.29890v1 Announce Type: new Abstract: Biomedical Named Entity Recognition (NER) is fundamental to healthcare AI applications, including clinical decision support and medical information ext...
arXiv:2606. 24102v1 Announce Type: cross Abstract: Most electronic health record (EHR) foundation models encode clinical events as discrete event tokens from a fixed vocabulary and therefore cannot directly represent events containing unseen concepts or new combinations of concepts and attributes such as numeric values.
The paper introduces CAST, a concept-guided artifact suppression tuning framework that uses sparse autoencoders to identify and suppress note-specific artifacts in clinical language models. CAST labels latent features with an LLM-assisted pipeline and ICD‑10 constraints, then fine‑tunes the model while providing post‑hoc per‑concept attributions for auditability. In experiments on MIMIC‑IV discharge‑note mortality prediction, CAST outperforms standard fine‑tuned encoders and competes with strong LLM baselines while offering a feature‑level audit trail of clinical concepts and suppressed artifacts.
This thesis explores how to select and adapt NLP models for global health literature when annotated data and computational resources are scarce. It compares skip‑gram word2vec models trained on increasingly large specialized corpora with BioWordVec for semantic tag discovery, finding that larger coverage does not always yield more useful domain associations. The study also evaluates convolutional spaCy models versus a RoBERTa transformer for named entity recognition, noting a trade‑off between higher F1 scores and longer inference time, and investigates MiniLM few‑shot versus BART‑MNLI zero‑shot classification for multi‑label topic classification, highlighting practical constraints of inference cost. "whyItMatters":"The work provides empirical guidance on balancing model accuracy and resource demands for building knowledge systems in low‑resource global health settings."
BioELX is a retrieve‑rerank framework for cross‑lingual biomedical entity linking that tackles two key problems: the English‑biased UMLS alias training data and the degradation caused by naïvely adding context. It fine‑tunes SapBERT_multi with Wikidata‑derived cross‑lingual alias supervision to create shared concept neighborhoods, and then reranks candidates using pretrained LLMs with mention‑anchored prompting to focus on the target mention. Experiments demonstrate state‑of‑the‑art performance on four benchmarks, improving Recall@1 by 4.8–18.2 percentage points without task‑specific annotations.
The paper introduces Distilled Rapid Embedding Transfer (DRET), a parameter‑efficient method that injects biomedical domain knowledge from large specialized models into a smaller general‑purpose model without retraining on the original specialized corpora. DRET evolves through iterative strategies—tokenizer‑merge (DRET 1.x), hybrid embedding averaging (DRET 2.0), priority‑based embedding transfer (DRET 3.x), and further refinements (DRET 4.x)—and demonstrates that a 66‑million‑parameter DistilBERT can achieve competitive or superior performance on token‑level PICO classification compared to much larger models, while remaining lightweight. The authors validate the embedding‑level transfer with cosine similarity, semantic‑shift, and t‑SNE analyses, highlighting DRET’s potential for scalable, resource‑efficient biomedical text mining.
The study investigates whether cross‑lingual clinical annotation projection can be treated as a constrained text‑generation task that preserves the original text while inserting entity tags. Using a workflow that embeds tags directly into immutable target‑language text and then validates them deterministically, the authors evaluated this approach against supervised candidate‑span projection and hybrid ML‑LLM refinement across six languages. Results show that direct LLM projection, particularly with GLM 5.2 and Gemma4:31B, achieves the highest strict F1 scores (up to 0.9201) and outperforms previous methods by 0.0564–0.1512, producing over 55,000 grounded mentions with accurate offsets.
arXiv:2607. 15447v1 Announce Type: new Abstract: Recent research in clinical machine learning, focusing on outcome predictions in intensive care unit (ICU), has shifted from bespoke supervised models to foundation models, utilising modern representation learning methods.
arXiv:2606. 28798v1 Announce Type: new Abstract: Objective: ICD codes are central to reimbursement, research, and population health surveillance, yet automated coding systems often struggle to integrate diagnostic signals from both clinical narratives and structured electronic health record (EHR) variables.
The paper introduces a benchmark for recommending SNOMED CT concepts from masked clinical contexts, using data from the SNOMED CT Entity Linking Challenge v1.2.1 derived from MIMIC-IV-Note. It evaluates several methods—including a popularity baseline, sparse TF‑IDF prototypes, dense embeddings, and retrieval‑augmented hybrids—finding that sparse TF‑IDF achieves the best performance with Recall@1 of 14.81% and Recall@10 of 33.43%. The study highlights that concept frequency and lexical context strongly influence recommendation quality, with many test pairs involving concepts unseen during training.