9 papers
Structural Alignment Improves Graph Test-Time Adaptation
Hans Hao-Hsun Hsu, Shikun Liu, Han Zhao +1
Graph-based learning excels at capturing interaction patterns in diverse domains like recommendation, fraud detection, and particle physics. However, its performance often degrades…
A Unified Theory of Random Projection for Influence Functions
Pingbang Hu, Yuzheng Hu, Jiaqi W. Ma +1
Influence functions and related data attribution scores take the form of , where is a curvature operator. In modern overparametrized models,…
FM SO.P: A Progressive Task Mixture Framework with Automatic Evaluation for Cross-Domain SOP Understanding
Siyuan Huang, Ziyu Wang, Chao Pan +1
Standard Operating Procedures (SOPs) are critical for enterprise operations, yet existing language models struggle with SOP understanding and cross-domain generalization. Current m…
Efficient Utility-Preserving Machine Unlearning with Implicit Gradient Surgery
Shiji Zhou, Tianbai Yu, Zhi Zhang +4
Machine unlearning (MU) aims to efficiently remove sensitive or harmful memory from a pre-trained model. The key challenge is to balance the potential tradeoff between unlearning e…
MiCRo: Mixture Modeling and Context-aware Routing for Personalized Preference Learning
Jingyan Shen, Jiarui Yao, Rui Yang +5
Reward modeling is a key step in building safe foundation models when applying reinforcement learning from human feedback (RLHF) to align Large Language Models (LLMs). However, rew…
Moment Alignment: Unifying Gradient and Hessian Matching for Domain Generalization
Yuen Chen, Haozhe Si, Guojun Zhang +1
Domain generalization (DG) seeks to develop models that generalize well to unseen target domains, addressing the prevalent issue of distribution shifts in real-world applications.…