LLM条件控制中的效果与流畅性权衡研究
On the Effectiveness-Fluency Trade-Off in LLM Conditioning: A Systematic Study
苹果团队研究了LLM条件控制中的效果与流畅性权衡，发现高效方法常牺牲流畅性。
苹果公司研究团队系统研究了大型语言模型的条件控制方法。研究聚焦概念注入和移除两种场景，评估多种 conditioning 技术。研究发现高效引导方法常以牺牲流畅性为代价实现条件控制。该研究揭示了当前方法在效果与质量间的权衡关系。
On the Effectiveness-Fluency Trade-Off in LLM Conditioning: A Systematic Study
Controlling the output of Large Language Models (LLMs) is a central challenge for their reliable deployment, yet a clear understanding of the involved trade-offs remains elusive. Current approaches to conditioning are often evaluated with a narrow focus on their effectiveness at injecting or removing a target concept, neglecting generation quality. We systematically investigate a range of conditioning methods in both injection and removal scenarios. We find that efficient steering methods frequently achieve conditioning at a steep cost to fluency. Furthermore, we identify a critical yet…
