SeMi-LoRA: Enhancing Low-Rank Adaptation via Separation and Mixing

Abstract

Low-Rank Adaptation (LoRA) has become a standard paradigm for parameter-efficient fine-tuning (PEFT). However, its low-rank constraint can limit its capacity to express complex weight updates, leaving a performance gap compared with full fine-tuning. Existing extensions improve expressivity, but they often sacrifice parameter mergeability or rely on sparse, block-diagonal updates that restrict global information flow. We propose SeMi-LoRA (Separation and Mixing LoRA), a novel framework that enables complete high-rank updates while preserving mergeability. SeMi-LoRA decomposes adaptation into channel-wise compression followed by two-stage latent mixing, namely Rank Mix and Channel Mix. Our structural analysis shows that, under a fixed base rank, the rank capacity of the update increases with the number of channels and supports complete updates rather than partial sparse ones. Extensive experiments on commonsense reasoning, mathematical reasoning, natural language understanding, and dialogue generation benchmarks demonstrate that SeMi-LoRA consistently outperforms strong PEFT baselines with favorable parameter efficiency.