Performance Analysis
Semi-Synthetic Transformers for Evaluating Mechanistic Interpretability Techniques
Training (IIT) which we call Strict IIT (SIIT). SIIT models maintain Tracr's original circuit while being more realistic.
ConStat: Performance-Based Contamination Detection in Large Language Models Jasper Dekoninck 1, Mark Niklas Müller 1,2, Martin Vechev 1 Department of Computer Science 1
Public benchmarks play an essential role in the evaluation of large language models.
LEACE: Perfect linear concept erasure in closed form
Concept erasure aims to remove specified features from a representation. It can improve fairness (e.g.
Lips Are Lying: Spotting the Temporal Inconsistency between Audio and Visual in Lip-Syncing DeepFakes
Our preliminary experiments have shown that the effectiveness of the existing methods often drastically decrease or even fail when tackling lip-syncing videos.
