17 citations · 26 across the 16 of their papers we have counts for
18 papers
Conformal Prediction with Paraphrase-Aware Scoring for LLM Uncertainty Quantification
Jiayi Xin, Evan Qiang, Zihan Zhu +3
Uncertainty quantification (UQ) for large language models (LLMs) aims to provide reliable measures of predictive confidence, yet current methods are often unstable under meaning-pr…
Optimal Watermark Localization in Mixed-Source Large Language Model Texts
Jose H. Blanchet, T. Tony Cai, Xiang Li +3
Watermarking provides a principled way to authenticate text generated by large language models (LLMs). In practice, however, the final text may be mixed-source, with watermark evid…
CurveRL: Principled Distribution-Aware Context Reweighting for LLM Reasoning
Ke Sun, Yizhou Zhao, Jiayi Xin +2
Context or prompt-level reweighting has emerged as a central algorithmic lever in Reinforcement Learning with Verified Rewards (RLVR) for improving the reasoning capability of larg…
Improving the Trade-off Between Watermark Strength and Speculative Sampling Efficiency for Language Models
Weiqing He, Xiang Li, Li Shen +2
Watermarking is a principled approach for tracing the provenance of large language model (LLM) outputs, but its deployment in practice is hindered by inference inefficiency. Specul…
Robust Spectral Watermark for Synthetic Tabular Data
Yizhou Zhao, Xiang Li, Peter Song +2
The rise of generative AI has enabled the production of high-fidelity synthetic tabular data across fields such as healthcare, finance, and public policy, raising growing concerns…
Optimal Detection for Language Watermarks with Pseudorandom Collision
T. Tony Cai, Xiang Li, Qi Long +2
Text watermarking plays a crucial role in ensuring the traceability and accountability of large language model (LLM) outputs and mitigating misuse. While promising, most existing m…
On the Empirical Power of Goodness-of-Fit Tests in Watermark Detection
Weiqing He, Xiang Li, Tianqi Shang +3
Large language models (LLMs) raise concerns about content authenticity and integrity because they can generate human-like text at scale. Text watermarks, which embed detectable sta…
Theoretical Tensions in RLHF: Reconciling Empirical Success with Inconsistencies in Social Choice Theory
Jiancong Xiao, Zhekun Shi, Kaizhao Liu +2
Despite its empirical success, Reinforcement Learning from Human Feedback (RLHF) has been shown to violate almost all the fundamental axioms in social choice theory -- such as majo…
Optimal Estimation of Watermark Proportions in Hybrid AI-Human Texts
Xiang Li, Garrett Wen, Weiqing He +3
Text watermarks in large language models (LLMs) are an increasingly important tool for detecting synthetic text and distinguishing human-written content from LLM-generated text. Wh…
Evaluating the Unseen Capabilities: How Many Theorems Do LLMs Know?
Xiang Li, Jiayi Xin, Qi Long +1
Accurate evaluation of large language models (LLMs) is crucial for understanding their capabilities and guiding their development. However, current evaluations often inconsistently…
Fundamental Limits of Game-Theoretic LLM Alignment: Smith Consistency and Preference Matching
Zhekun Shi, Kaizhao Liu, Qi Long +2
Nash Learning from Human Feedback is a game-theoretic framework for aligning large language models (LLMs) with human preferences by modeling learning as a two-player zero-sum game.…
Riemannian EXTRA: Communication-efficient decentralized optimization over compact submanifolds with data heterogeneity
Jiayuan Wu, Zhanwang Deng, Jiang Hu +2
We consider decentralized optimization over a compact Riemannian submanifold in a network of agents, where each agent holds a smooth, nonconvex local objective defined by its p…
PolarGrad: A Class of Matrix-Gradient Optimizers from a Unifying Preconditioning Perspective
Tim Tsz-Kit Lau, Qi Long, Weijie Su
The ever-growing scale of deep learning models and training data underscores the critical importance of efficient optimization methods. While preconditioned gradient methods such a…
Statistical Impossibility and Possibility of Aligning LLMs with Human Preferences: From Condorcet Paradox to Nash Equilibrium
Kaizhao Liu, Qi Long, Zhekun Shi +2
Aligning large language models (LLMs) with diverse human preferences is critical for ensuring fairness and informed outcomes when deploying these models for decision-making. In thi…
Robust Detection of Watermarks for Large Language Models Under Human Edits
Xiang Li, Feng Ruan, Huiyuan Wang +2
Watermarking has offered an effective approach to distinguishing text generated by large language models (LLMs) from human-written text. However, the pervasive presence of human ed…
Debiasing Watermarks for Large Language Models via Maximal Coupling
Yangxinyu Xie, Xiang Li, Tanwi Mallick +2
Watermarking language models is essential for distinguishing between human and machine-generated text and thus maintaining the integrity and trustworthiness of digital communicatio…
On the Algorithmic Bias of Aligning Large Language Models with RLHF: Preference Collapse and Matching Regularization
Jiancong Xiao, Ziniu Li, Xingyu Xie +4
Accurately aligning large language models (LLMs) with human preferences is crucial for informing fair, economically sound, and statistically efficient decision-making processes. Ho…
A Statistical Framework of Watermarks for Large Language Models: Pivot, Detection Efficiency and Optimal Rules
Xiang Li, Feng Ruan, Huiyuan Wang +2
Since ChatGPT was introduced in November 2022, embedding (nearly) unnoticeable statistical signals into text generated by large language models (LLMs), also known as watermarking,…