Jiachen Sun

Jiachen Sun

Auto-generated profile, based on publicly available information

Jiachen Sun

University of Michigan

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
15 October 2018
Verified
Conference Paper
Open access
PDF file
ACAnadi Chaman
Anadi Chaman
JWJiaming Wang
Jiaming Wang
JSJiachen Sun
Jiachen Sun
HHHaitham Hassanieh
Haitham Hassanieh
RCRomit Roy Choudhury
Romit Roy Choudhury

4 November 2024
Verified
Conference Paper
Open access
RZRuiyang Zhu
Ruiyang Zhu
XZXiao Zhu
Xiao Zhu
AZAnlan Zhang
Anlan Zhang
XZXumiao Zhang
Xumiao Zhang
JSJiachen Sun
Jiachen Sun
MLMyungjin Lee
Myungjin Lee

28 October 2022
Verified
Book Chapter
Closed access
PDF file
JSJiachen Sun
Jiachen Sun
AMAkshay Mehra
Akshay Mehra
BKBhavya Kailkhura
Bhavya Kailkhura
PCPin-Yu Chen
Pin-Yu Chen
DHDan Hendrycks
Dan Hendrycks
ZMZ. Morley Mao
Z. Morley Mao

26 October 2023
Verified
Conference Paper
Open access
JSJiachen Sun
Jiachen Sun
MIMark Ibrahim
Mark Ibrahim
MHMelissa Hall
Melissa Hall
IEIvan Evtimov
Ivan Evtimov
ZMZ. Morley Mao
Z. Morley Mao
CHCaner Hazirbas
Caner Hazirbas
Textual prompt tuning has demonstrated significant performance improvements in adapting natural language processing models to a variety of downstream tasks by treating hand-engineered prompts as trainable parameters. Inspired by the success of textual prompting, several studies have investigated the efficacy of visual prompt tuning. In this work, we present Visual Prompt Adaptation (VPA), the first framework that generalizes visual prompting with test-time adaptation. VPA introduces a small number of learnable tokens, enabling fully test-time and storage-efficient adaptation without necessitating source-domain information. We examine our VPA design under diverse adaptation settings, encompassing single-image, batched-image, and pseudo-label adaptation. We evaluate VPA on multiple tasks, including out-of-distribution (OOD) generalization, corruption robustness, and domain adaptation. Experimental results reveal that VPA effectively enhances OOD generalization by 3.3% across various models, surpassing previous test-time approaches. Furthermore, we show that VPA improves corruption robustness by 6.5% compared to strong baselines. Finally, we demonstrate that VPA also boosts domain adaptation performance by relatively 5.2%. Our VPA also exhibits marked effectiveness in improving the robustness of zero-shot recognition for vision-language models.

1 April 2019
Verified
Conference Paper
Closed access
YLYang Li
Yang Li
JSJiachen Sun
Jiachen Sun
WHWenguang Huang
Wenguang Huang
XTXiaohua Tian
Xiaohua Tian
In this paper, we propose a tree modeling-based data mining method to detect anomalies from crowdsourced network data. We design an algorithm to extract potential network anomalies from decision trees. Moreover, we propose a criteria to evaluate the severity of anomaly in terms of three factors: standard deviation, weight sum and impurity decrease. To enhance generalization performance, we randomly generate sample subspace of the original dataset as the input for each subtree and compact detected anomalies from all subtrees. We carry out experiments based on the crowdsourced network measurement dataset containing five million samples, which contains round trip time (RTT) from more than 5,000 users. Experiments show that the proposed method can effectively detect high-latency network anomalies. Moreover, the random forest-based approach can achieve an improvement of approximately 25% of generalization performance compared to the single decision tree approach.

3 November 2024
Verified
Book Chapter
Closed access
PDF file
HZHaizhong Zheng
Haizhong Zheng
JSJiachen Sun
Jiachen Sun
SWShutong Wu
Shutong Wu
BKBhavya Kailkhura
Bhavya Kailkhura
ZMZ. Morley Mao
Z. Morley Mao
APAtul Prakash
Atul Prakash