Adapting large language models to multiple tasks can cause cross-skill interference, where improvements for one skill degrade another. While methods such as LoRA impose orthogonality constraints at the weight level, they do not fully address interference in hidden-state representations. We propose Compositional Subspace Representation Fine-tuning (CS-ReFT), a novel representation-based approach that learns multiple orthonormal subspace transformations, each specializing in a distinct skill, and composes them via a lightweight router. By isolating these subspace edits in the hidden state, rather than weight matrices, CS-ReFT prevents cross-task conflicts more effectively. On the AlpacaEval benchmark, applying CS-ReFT to Llama-2-7B achieves a 93.94% win rate, surpassing GPT-3.5 Turbo (86.30%) while requiring only 0.0098% of model parameters. These findings show that specialized representation edits, composed via a simple router, significantly enhance multi-task instruction following with minimal overhead.

本研究解决了大型语言模型在多任务适应中出现的交叉技能干扰问题，提出了一种名为组合子空间表示微调（CS-ReFT）的新方法，该方法通过学习多个正交子空间变换来专注于不同技能，并通过轻量级路由器进行组合。研究表明，CS-ReFT在AlpacaEval基准测试中表现优异，有效提高了多任务指令执行的能力，且模型参数需求极低。

适应性大型语言模型的组合子空间表示微调