conftrace_

white-box method

2 papers

Explore in graph

Co-occurring keywords · papers mentioning both

distributional distance (1) high-frequency artifact (1) embedding distance (1) faithfulness (1) forgrad (1) gradient signal (1) large language model (1) embedding analysis (1) taylor expansion (1) gradient information (1)

Papers

Embedding and Gradient Say Wrong: A White-Box Method for Hallucination Detection EMNLP 2024 Saliency strikes back: How filtering out high frequencies improves white-box explanations ICML 2024