SALSA: Speech Aware LLM Adaptation via Learned Steering Activation Vectors 事件

SALSA: Speech Aware LLM Adaptation via Learned Steering Activation Vectors arXiv:2606.00460v1 Announce Type: new Abstract: Speech-aware large language models often generalize poorly to out-of-domain settings. We propose SALSA (Speech-Aware LLM Adaptation via Learned Steering Activations), a lightweight adaptation method that learns layer-wise steering vectors. Unlike commonly used steering approaches that rely on contrastive activation differences, SALSA directly optimizes steering vectors usin