Work performed while at Stanford University.

We illustrate these bounds in the logistic regression and neural network settings. In contrast, we establish settings where learning one representation for all tasks (i.e. using a "frozen representation" objective) fails. Notably, any such algorithm cannot outperform directly learning the target task with no other information, in the worst case.

Releasing open-source large language models (LLMs) presents a dual-use risk since bad actors can easily fine-tune these models for harmful purposes. Even without the open release of weights, weight stealing and fine-tuning APIs make closed models vulnerable to harmful fine-tuning attacks (HFAs).

Efficiency: The overall reward can be allocated to all players in the game,i.e. This section provides more details about multi-order interactions [8] in Section 3.3 of the paper. The multi-order interaction satisfies axioms oflinearity, nullity, commutativity, symmetry, and efficiency[8],asfollows. This study was done under the supervision of Dr. Quanshi Zhang. This section provides more details about the use of the ShapeNet part dataset in the paper.

This paper focuses on the intersection of the above two directions,i.e.