@inproceedings{ondov-etal-2026-coreelm,
    title = "{C}ore{ELM}: An Open-Source Framework for Aligning Large Language Models to Embedding Spaces",
    author = "Ondov, Brian  and
      Chang, Chia-Hsuan  and
      Zhou, Yujia  and
      Giuffr{\`e}, Mauro  and
      Xu, Hua",
    editor = "Demner-Fushman, Dina  and
      Ananiadou, Sophia  and
      Roberts, Kirk  and
      Tsujii, Junichi",
    booktitle = "{B}io{NLP} 2026",
    month = jul,
    year = "2026",
    address = "San Diego, California",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl-workshops/2026.bionlp-1.15/",
    pages = "156--180",
    ISBN = "979-8-89176-434-7",
    abstract = "Text embeddings have become an essential part of a variety of language applications. However, methods for interpreting, exploring and reversing embedding spaces are limited, reducing transparency and precluding potentially valuable generative use cases. In this work, we develop an open-source, domain-agnostic framework for aligning Large Language Models to embedding spaces using the recently reported Embedding Language Model (ELM) method. We demonstrate our framework by training models to recover, summarize, and compare clinical trial abstracts from embeddings alone. In addition to inverting embeddings back to text more reliably than existing methods, our models can decode novel, interpolated embeddings into new clinical trial abstracts that human experts cannot distinguish from real ones. We further show that these generated abstracts are responsive to moving embeddings along concept vectors for age and sex of study subjects. Our public ELM implementation and experimental results will aid the alignment of Large Language Models to embedding spaces in the biomedical domain and beyond."
}