LoTR: Low Tensor Rank Weight Adaptation
This is an incremental improvement for researchers and practitioners working on efficient fine-tuning of large language models.
The paper tackles the problem of parameter-efficient fine-tuning for large language models by introducing LoTR, a method that uses tensor decomposition to represent gradient updates, achieving better parameter efficiency than LoRA, especially for deep models, with the core tensor being arbitrarily small for cheap and fast fine-tuning.
In this paper we generalize and extend an idea of low-rank adaptation (LoRA) of large language models (LLMs) based on Transformer architecture. Widely used LoRA-like methods of fine-tuning LLMs are based on matrix factorization of gradient update. We introduce LoTR, a novel approach for parameter-efficient fine-tuning of LLMs which represents a gradient update to parameters in a form of tensor decomposition. Low-rank adapter for each layer is constructed as a product of three matrices, and tensor structure arises from sharing left and right multipliers of this product among layers. Simultaneous compression of a sequence of layers with low-rank tensor representation allows LoTR to archive even better parameter efficiency then LoRA especially for deep models. Moreover, the core tensor does not depend on original weight dimension and can be made arbitrary small, which allows for extremely cheap and fast downstream fine-tuning.