Abstract
While domain-specific Large Language Models (LLMs) have encoded vast biomedical knowledge, their limited context windows often hinder a deep understanding of nuanced relationships within and across texts. To address this limitation, we introduce BioBigBird, a bidirectional language model pre-trained on extensive biomedical literature and clinical data, specifically designed to handle long-range dependencies. BioBigBird leverages a sparse attention mechanism to process sequences up to 4096 tokens, and its training incorporates a multi-stage process to mitigate noise from the large-scale pre-training corpus. We further enhance its performance by employing a multi-task learning (MTL) framework that jointly optimizes for Named Entity Recognition and Relation Extraction. Comprehensive evaluations on the BLURB benchmark reveal that our MTL-enhanced BioBigBird achieves highly competitive results against state-of-the-art models. Our work contributes an effective methodology for developing powerful, long-context language models for specialized domains, demonstrating the value of extended sequence processing for complex text analysis. Our models are publicly available at https://huggingface.co/collections/bisectgroup/biobigbird.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Balaji, R., S, P. K., Gupta, V., N, S., Sridhar, K., & Bhatt, N. (2026). BioBigBird: A Sparse Attention Model for Long-Range Dependency Processing in Biomedical Text. https://omanscience.com/en/articles/biobigbird-a-sparse-attention-model-for-long-range-dependency-processing-in-biomedical-text
MLA 9
Balaji, Roshan, et al. "BioBigBird: A Sparse Attention Model for Long-Range Dependency Processing in Biomedical Text." https://omanscience.com/en/articles/biobigbird-a-sparse-attention-model-for-long-range-dependency-processing-in-biomedical-text.
Chicago (author–date)
Balaji, Roshan, Pavan Kumar S, Vasudev Gupta, Sreejith N, Keerthana Sridhar, and Nirav Bhatt. 2026. "BioBigBird: A Sparse Attention Model for Long-Range Dependency Processing in Biomedical Text." https://omanscience.com/en/articles/biobigbird-a-sparse-attention-model-for-long-range-dependency-processing-in-biomedical-text.
Harvard
Balaji, R., S, P. K., Gupta, V., N, S., Sridhar, K. and Bhatt, N. (2026) 'BioBigBird: A Sparse Attention Model for Long-Range Dependency Processing in Biomedical Text', Available at: https://omanscience.com/en/articles/biobigbird-a-sparse-attention-model-for-long-range-dependency-processing-in-biomedical-text.
Vancouver
Balaji R, S PK, Gupta V, N S, Sridhar K, Bhatt N. BioBigBird: A Sparse Attention Model for Long-Range Dependency Processing in Biomedical Text. https://omanscience.com/en/articles/biobigbird-a-sparse-attention-model-for-long-range-dependency-processing-in-biomedical-text
IEEE
R. Balaji, P. K. S, V. Gupta, S. N, K. Sridhar, and N. Bhatt, "BioBigBird: A Sparse Attention Model for Long-Range Dependency Processing in Biomedical Text," https://omanscience.com/en/articles/biobigbird-a-sparse-attention-model-for-long-range-dependency-processing-in-biomedical-text.