Weijie J Su

Improving the Trade-off Between Watermark Strength and Speculative Sampling Efficiency for Language Models

TAB-DRW: A DFT-based Robust Watermark for Generative Tabular Data

Fundamental Limits of Game-Theoretic LLM Alignment: Smith Consistency and Preference Matching

Paraphrase-Robust Conformal Prediction for Reliable LLM Uncertainty Quantification

LoRA Provably Reduces Forgetting and Enables Adapter Merging in Multiclass Linear Classification

A Fine-Grained Analysis of Pure Semantic Preference Alignment in Large Language Models

On the Empirical Power of Goodness-of-Fit Tests in Watermark Detection

Mitigating the Privacy–Utility Trade-off in Decentralized Federated Learning via f-Differential Privacy

Magnetic Preference Optimization: Achieving Last-iterate Convergence for Language Model Alignment

Fundamental Limits of Game-Theoretic LLM Alignment: Smith Consistency and Preference Matching

Fine-Tuning Attention Modules Only: Enhancing Weight Disentanglement in Task Arithmetic

Restoring Calibration for Aligned Large Language Models: A Calibration-Aware Fine-Tuning Approach