3 papers indexed here
We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.
Not the right person? Other researchers publish under this name.
Activation steering offers an effective and efficient way to modify an LLM's behavior at inference time. However, it remains unclear how to apply steering such that the signal is consistently strong and the steering strength has a universally interpretable meaning. We address this first by identifying the most steerabl...
This work examines how program inspection provides insights beyond accuracy scores and helps guide training and assessed training data improvements informed by program inspection and failure analysis, finding gains in sequence accuracy on several tasks.
This work releases a version of TruthfulQA with surface-feature leakage reduced close to chance and provides a mechanism, Audit-Prune, so that the datasets can be cleaned before release.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
