Deep Noir 框架可自动发现 LLM 推理时的最优引导参数
Deep Noir: Autonomous Steering Discovery via Architectural Chronometry in Transformer Models
Deep Noir 这个框架挺有意思，能自动发现 LLM 推理时的最优引导参数，不用手动调参数了，效果还挺好。
Deep Noir 是一个框架，它使用 Logit Lens 收敛和因果头级归因来自主发现 LLM 推理时的最优引导参数。在 1B 模型上，它在垃圾邮件检测任务上实现了 16.7 个百分点的提升（标准差 4.7，39 次运行），在 7-9B 模型上，它在四个架构上实现了 21 到 42 个百分点的提升。在 SST-2 情感分析任务上，它实现了 13.1 个百分点的提升，且无需修改代码。机械因果解释使干预点能够跨任务和架构泛化。
Deep Noir: Autonomous Steering Discovery via Architectural Chronometry in Transformer Models
Activation steering modifies LLM behavior at inference time, but identifying where and how strongly to steer remains manual. We introduce Deep Noir, a framework that uses Logit Lens convergence and causal head-level attribution to autonomously discover optimal steering parameters. Across three scales (1B x 3, 2-3B x 2, and 7-9B x 4), our engine achieves 16.7 percentage-point improvement on spam at 1B (standard deviation 4.7; 39 runs), with gains increasing to 21 to 42 percentage points at 7-9B across four architectures. On SST-2 sentiment, it achieves a 13.1 percentage-point improvement with zero code changes. Mechanistic grounding enables automated discovery of intervention points that generalize across tasks and architectures. On sentiment, RepE without head masking fails to improve over baseline, while Deep Noir improves all models (p less than 0.01). We further show that steering creates a predictable prompt-injection attack surface whose vulnerability increases monotonically with steering magnitude. This finding is relevant to agent systems deploying steered classifiers.
