Qi Long

Professor, University of Pennsylvania

3 papers at NeurIPS 2025

Homepage· OpenReview· Semantic Scholar· Google Scholar

Poster Session 1

Wednesday, December 3, 2025 · 11:00 AM → 2:00 PM

BetaConform

: Efficient MAP Estimation of LLM Ensemble Judgment Performance with Prior Transfer

#3519 · Huaizhi Qu, Inyoung Choi, Zhen Tan, Song Wang, Sukwon Yun, Qi Long, Faizan Siddiqui, Kwonjoon Lee, Tianlong Chen

We present an efficient judgmetn distribution estimation method for LLM ensembles.

Poster Session 2

2 papers

Wednesday, December 3, 2025 · 4:30 PM → 7:30 PM

Exhibit Hall C,D,E

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

#1211 Spotlight · Xiang Li, Chendi Wang, Buxin Su, Qi Long, Weijie J Su

We develop a fine-grained f-DP analysis for decentralized federated learning, improving privacy–utility trade-offs under random walk communication and extending to settings with dependent noise.

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

#3716 Spotlight · Weiqing He, Xiang Li, Tianqi Shang, Li Shen, Weijie J Su, Qi Long

Introduce and apply Goodness-of-Fit (GoF) tests to watermark detection, and explore their effectiveness across diverse conditions.