DNACHUNKER: Learnable Tokenization for DNA Language Models

Abstract

DNA language models are increasingly used to represent genomic sequence, yet their effectiveness depends critically on how raw nucleotides are converted into model inputs. Unlike natural language, DNA offers no canonical boundaries, making fixed tokenizations a brittle design choice under shifts, indels, and local repeats. We introduce \modelname{}, a masked DNA language model that incorporates a learnable adaptive segmentation module to produce context-dependent, variable-length units. Building on a dynamic segmentation procedure, \modelname{} learns to allocate finer granularity to functionally enriched regions while compressing repetitive or redundant sequence. We pre-train \modelname{} on the human reference genome (HG38) and evaluate it on the Nucleotide Transformer and Genomic Benchmarks, where it consistently improves over strong fixed-tokenization baselines. Further analyses and ablations indicate that the learned segmentation is structured rather than incidental: the model preferentially uses shorter units around promoters and exons, and longer units in repetitive regions, yielding representations that are both mutation-resilient and biologically-informed.