activity
20232026
collaborators

19 papers

cs.LG2026

AdaFlash: Adaptive Speculative Decoding via On-Policy Distilled Diffusion Drafters

Yu-Yang Qian, Hao-Cong Wu, Chen Chen +4

Speculative decoding, in which a lightweight draft model first generates a draft sequence that is then verified by the target model, has become a prevalent paradigm for acceleratin…

cs.LG2026

Nonlinear Bandit

Tianshuo Zheng, Ting Wu, Zhi-Hua Zhou +1

In this paper we first study the problem of generalized linear bandit (GLB) under heavy-tailed noise. The characteristics of heavy-tailed distributions are widely observed in real-…

cs.LG2026

Online Learning with Gradient-Variation Interval Regret

Yan-Feng Xie, Shuche Wang, Peng Zhao +1

This paper investigates non-stationary online learning using the metric of interval regret, which requires an online algorithm to perform well over every time interval. We propose…

cs.LG2026

Robust Length Prediction: A Perspective from Heavy-Tailed Prompt-Conditioned Distributions

Jing Wang, Yu-Yang Qian, Ke Xue +3

Output-length prediction is important for efficient LLM serving, as it directly affects batching, memory reservation, and scheduling. For prompt-only length prediction, most existi…

cs.LG2026

Dynamic Regret via Discounted-to-Dynamic Reduction with Applications to Curved Losses and Adam Optimizer

Yan-Feng Xie, Yu-Jie Zhang, Peng Zhao +1

We study dynamic regret minimization in non-stationary online learning, with a primary focus on follow-the-regularized-leader (FTRL) methods. FTRL is important for curved losses an…

cs.DB2026

AeroSketch: Near-Optimal Time Matrix Sketch Framework for Persistent, Sliding Window, and Distributed Streams

Hanyan Yin, Dongxie Wen, Jiajun Li +4

Many real-world matrix datasets arrive as high-throughput vector streams, making it impractical to store or process them in their entirety. To enable real-time analytics under limi…