Batched Low-Rank Adaptation of Foundation Models
This addresses a performance bottleneck for real-time serving to diverse users in scenarios requiring personalized adaptations, representing an incremental improvement over existing LoRA methods.
The paper tackles the problem of efficiently serving multiple task-specific adaptations in foundation models by introducing Fast LoRA (FLoRA), which allows batching heterogeneous requests with unique low-rank adaptation weights per input, achieving competitive results on MultiPL-E code generation across 8 languages and multilingual speech recognition across 6 languages.
Low-Rank Adaptation (LoRA) has recently gained attention for fine-tuning foundation models by incorporating trainable low-rank matrices, thereby reducing the number of trainable parameters. While LoRA offers numerous advantages, its applicability for real-time serving to a diverse and global user base is constrained by its incapability to handle multiple task-specific adapters efficiently. This imposes a performance bottleneck in scenarios requiring personalized, task-specific adaptations for each incoming request. To mitigate this constraint, we introduce Fast LoRA (FLoRA), a framework in which each input example in a minibatch can be associated with its unique low-rank adaptation weights, allowing for efficient batching of heterogeneous requests. We empirically demonstrate that FLoRA retains the performance merits of LoRA, showcasing competitive results on the MultiPL-E code generation benchmark spanning over 8 languages and a multilingual speech recognition task across 6 languages.