Inthis work, we show that the reason for this deficiency is that language models are biased to learn word co-occurrence statistics instead of true factual associations.

The work was conducted during the internship of Y uqi Chen and Y uchen Fang at Microsoft Research.

Recent methodologies in LLM self-training mostly rely on LLM generating responses and filtering those with correct output answers as training data. This approach often yields a low-quality fine-tuning training set (e.g., incorrect plans or intermediate reasoning).

Machine learning models are increasingly being deployed in real-world contexts. However, systematic studies on their transferability to specific and critical applications are underrepresented in the research literature. An important example is visual anomaly detection (V AD) for robotic power line inspection.

We investigated DNNs as encoding models of visual areas V1, V2, V4, and IT. We examined 32 visual deep neural networks.

Sharpness-aware minimization (SAM) is a recently proposed method that minimizes the sharpness of the training loss of a neural network.