Model Decides How to Tokenize: Adaptive DNA Sequence Tokenization with MxDNA
This addresses the challenge of adapting tokenization methods for DNA sequences in genomics, offering a novel approach that could impact various domains, though it appears incremental in improving existing methods.
The paper tackles the problem of suboptimal tokenization for DNA sequences in foundation models by introducing MxDNA, a framework where the model autonomously learns tokenization through gradient descent, achieving superior performance on benchmarks with less pretraining data and time.
Foundation models have made significant strides in understanding the genomic language of DNA sequences. However, previous models typically adopt the tokenization methods designed for natural language, which are unsuitable for DNA sequences due to their unique characteristics. In addition, the optimal approach to tokenize DNA remains largely under-explored, and may not be intuitively understood by humans even if discovered. To address these challenges, we introduce MxDNA, a novel framework where the model autonomously learns an effective DNA tokenization strategy through gradient decent. MxDNA employs a sparse Mixture of Convolution Experts coupled with a deformable convolution to model the tokenization process, with the discontinuous, overlapping, and ambiguous nature of meaningful genomic segments explicitly considered. On Nucleotide Transformer Benchmarks and Genomic Benchmarks, MxDNA demonstrates superior performance to existing methods with less pretraining data and time, highlighting its effectiveness. Finally, we show that MxDNA learns unique tokenization strategy distinct to those of previous methods and captures genomic functionalities at a token level during self-supervised pretraining. Our MxDNA aims to provide a new perspective on DNA tokenization, potentially offering broad applications in various domains and yielding profound insights.