UniSteer: Text-Guided Flow Matching in Activation Space for Versatile LLM Steering 事件

UniSteer: Text-Guided Flow Matching in Activation Space for Versatile LLM Steering arXiv:2605.30076v1 Announce Type: new Abstract: Activation-based control steers large language models (LLMs) by intervening on their internal representations during inference, and has emerged as an effective paradigm for controlling behaviors such as persona and style. However, existing methods often rely on fixed steering directions or task-specific intervention modules, making them difficult to adapt to fine-gr