activity
20232025
most citedA Theoretical Insight into Attack and Defense of Gradient Leakage in Transformer

3 citations · 6 across the 7 of their papers we have counts for

collaborators
Showing cs.LGShow all

6 papers · 1 filter

cs.LG2025

Theoretical Foundation of Flow-Based Time Series Generation: Provable Approximation, Generalization, and Efficiency

Jiangxuan Long, Zhao Song, Chiwun Yang

Recent studies suggest utilizing generative models instead of traditional auto-regressive algorithms for time series forecasting (TSF) tasks. These non-auto-regressive approaches i…

cs.LG2024

Unlocking the Theory Behind Scaling 1-Bit Neural Networks

Majid Daliri, Zhao Song, Chiwun Yang

Recently, 1-bit Large Language Models (LLMs) have emerged, showcasing an impressive combination of efficiency and performance that rivals traditional LLMs. Research by Wang et al.…

cs.LG2023★ 3 cited

A Theoretical Insight into Attack and Defense of Gradient Leakage in Transformer

Chenyang Li, Zhao Song, Weixin Wang +1

The Deep Leakage from Gradient (DLG) attack has emerged as a prevalent and highly effective method for extracting sensitive training data by inspecting exchanged gradients. This ap…

cs.LG2023

One Pass Streaming Algorithm for Super Long Token Attention Approximation in Sublinear Space

Raghav Addanki, Chenyang Li, Zhao Song +1

Attention computation takes both the time complexity of and the space complexity of simultaneously, which makes deploying Large Language Models (LLMs) in streamin…

cs.LG2023

An Automatic Learning Rate Schedule Algorithm for Achieving Faster Convergence and Steeper Descent

Zhao Song, Chiwun Yang

The delta-bar-delta algorithm is recognized as a learning rate adaptation technique that enhances the convergence speed of the training process in optimization by dynamically sched…

cs.LG2023★ 3 cited

Fine-tune Language Models to Approximate Unbiased In-context Learning

Timothy Chu, Zhao Song, Chiwun Yang

In-context learning (ICL) is an astonishing emergent ability of large language models (LLMs). By presenting a prompt that includes multiple input-output pairs as examples and intro…