Peilin Zhong

Researcher@Meta·美国·OpenReview

研究方向

Large Language Model · Differential Privacy · Parallel Algorithms · Streaming Algorithms · Sketching Algorithms · Clustering · High Dimensional Geometry · Graph Algorithms

Peilin Zhong

TNT: Improving Chunkwise Training for Test-Time Memorization

ATLAS: Learning to Optimally Memorize the Context at Test Time

Memory Caching: RNNs with Growing Memory

It's All Connected: A Journey Through Test-Time Memorization, Attentional Bias, Retention, and Online Optimization

Titans: Learning to Memorize at Test Time

PiKE: Adaptive Data Mixing for Large-Scale Multi-Task Learning Under Low Gradient Conflicts

Nearly-Linear Time and Massively Parallel Algorithms for $k$-anonymity

Retraining with Predicted Hard Labels Provably Increases Model Accuracy

MS-SSM: A Multi-Scale State Space Model for Efficient Sequence Modeling

Differentially Private Space-Efficient Algorithms for Counting Distinct Elements in the Turnstile Model

Maximum Coverage in Turnstile Streams with Applications to Fingerprinting Measures

Nested Learning: The Illusion of Deep Learning Architectures

Addax: Utilizing Zeroth-Order Gradients to Improve Memory Efficiency and Performance of SGD for Fine-Tuning Language Models

Maximum Coverage in Turnstile Streams with Applications to Fingerprinting Measures

Retraining with Predicted Hard Labels Provably Increases Model Accuracy