Research | Chulin Xie

2024

On Memorization of Large Language Models in Logical Reasoning

Chulin Xie, Yangsibo Huang, Chiyuan Zhang, Da Yu, Xinyun Chen, Bill Yuchen Lin, Bo Li, Badih Ghazi, and Ravi Kumar

Preprint 2024

[arXiv] [Code] [Website]
(α) Crosslingual Capabilities and Knowledge Barriers in Multilingual Large Language Models

Lynn Chua, Badih Ghazi, Yangsibo Huang, Pritish Kamath, Ravi Kumar, Pasin Manurangsi, Amer Sinha, Chulin Xie, and Chiyuan Zhang

Preprint 2024

[arXiv] [Code]
RedCode: Risky Code Execution and Generation Benchmark for Code Agents

Chengquan Guo*, Xun Liu*, Chulin Xie*, Andy Zhou, Yi Zeng, Zinan Lin, Dawn Song, and Bo Li

NeurIPS 2024

[PDF]
Differentially Private Synthetic Data via Foundation Model APIs 2: Text

Chulin Xie, Zinan Lin, Arturs Backurs, Sivakanth Gopi, Da Yu, Huseyin A Inan, Harsha Nori, Haotian Jiang, Huishuai Zhang, Yin Tat Lee, Bo Li, and Sergey Yekhanin

ICML 2024 (Spotlight)

[arXiv] [Code] [Poster] [Website]
Decoding Compressed Trust: Scrutinizing the Trustworthiness of Efficient LLMs Under Compression

Junyuan Hong, Jinhao Duan, Chenhui Zhang, Zhangheng Li, Chulin Xie, Kelsey Lieberman, James Diffenderfer, Brian Bartoldson, Ajay Jaiswal, Kaidi Xu, Bhavya Kailkhura, Dan Hendrycks, Dawn Song, Zhangyang Wang, and Bo Li

ICML 2024

[arXiv] [Code] [Website]
LLM-PBE: Assessing Data Privacy in Large Language Models

Qinbin Li*, Junyuan Hong*, Chulin Xie*, Jeffrey Tan, Rachel Xin, Junyi Hou, Xavier Yin, Zhun Wang, Dan Hendrycks, Zhangyang Wang, Bo Li, Bingsheng He, and Dawn Song

VLDB 2024 (Best Research Paper Nomination)

[arXiv] [Code] [Website]
Graph Chain-of-Thought: Augmenting Large Language Models by Reasoning on Graphs

Bowen Jin, Chulin Xie, Jiawei Zhang, Kashob Kumar Roy, Yu Zhang, Suhang Wang, Yu Meng, and Jiawei Han

ACL 2024

[arXiv] [Code]
PerAda: Parameter-Efficient Federated Learning Personalization with Generalization Guarantees

Chulin Xie, De-An Huang, Wenda Chu, Daguang Xu, Chaowei Xiao, Bo Li, and Anima Anandkumar

CVPR 2024

[Abs] [arXiv] [Code] [Poster]

In this paper, we propose PerAda, a parameter-efficient pFL framework that reduces communication and computational costs and exhibits superior generalization performance, especially under test-time distribution shifts. PerAda reduces the costs by leveraging the power of pretrained models and only updates and communicates a small number of additional parameters from adapters. PerAda has good generalization since it regularizes each client’s personalized adapter with a global adapter, while the global adapter uses knowledge distillation to aggregate generalized information from all clients. Theoretically, we provide generalization bounds to explain why PerAda improves generalization, and we prove its convergence to stationary points under non-convex settings. Empirically, PerAda demonstrates competitive personalized performance (+4.85% on CheXpert) and enables better out-of-distribution generalization (+5.23% on CIFAR-10-C) on different datasets across natural and medical domains compared with baselines, while only updating 12.6% of parameters per model based on the adapter.
FedSelect: Personalized Federated Learning with Customized Selection of Parameters for Fine-Tuning

Rishub Tamirisa, Chulin Xie, Wenxuan Bao, Andy Zhou, Ron Arel, and Aviv Shamsian

CVPR 2024

[Abs] [arXiv] [Code]

In this work, we propose FedSelect, a novel PFL algorithm inspired by the iterative subnetwork discovery procedure used for the Lottery Ticket Hypothesis. FedSelect incrementally expands subnetworks to personalize client parameters, concurrently conducting global aggregations on the remaining parameters. This approach enables the personalization of both client parameters and subnetwork structure during the training process. Finally, we show that FedSelect outperforms recent state-of-the-art PFL algorithms under challenging client data heterogeneity settings and demonstrates robustness to various real-world distributional shifts.
Improving Privacy-Preserving Vertical Federated Learning by Efficient Communication with ADMM

Chulin Xie, Pin-Yu Chen, Qinbin Li, Arash Nourian, Ce Zhang, and Bo Li

SaTML 2024

[Abs] [arXiv] [PDF] [Poster]

In this paper, we propose an efficient VFL with multiple linear heads (VIM) framework, where each head corresponds to local clients by taking the separate contribution of each client into account. In addition, we propose an Alternating Direction Method of Multipliers (ADMM)-based method to solve our optimization problem, which reduces the communication cost by allowing multiple local updates in each step, and thus leads to better performance under differential privacy.
Ring-A-Bell! How Reliable are Concept Removal Methods for Diffusion Models?

Yu-Lin Tsai, Chia-Yi Hsu, Chulin Xie, Chih-Hsun Lin, Jia-You Chen, Bo Li, Pin-Yu Chen, Chia-Mu Yu, and Chun-Ying Huang

ICLR 2024

[Abs] [arXiv] [PDF] [Code]

Diffusion models for text-to-image (T2I) synthesis, such as Stable Diffusion (SD), have recently demonstrated exceptional capabilities for generating high-quality content. However, this progress has raised several concerns of potential misuse, particularly in creating copyrighted, prohibited, and restricted content, or NSFW (not safe for work) images. While efforts have been made to mitigate such problems, either by implementing a safety filter at the evaluation stage or by fine-tuning models to eliminate undesirable concepts or styles, the effectiveness of these safety measures in dealing with a wide range of prompts remains largely unexplored. In this work, we aim to investigate these safety mechanisms by proposing one novel concept retrieval algorithm for evaluation. We introduce Ring-A-Bell, a model-agnostic red-teaming tool for T2I diffusion models, where the whole evaluation can be prepared in advance without prior knowledge of the target model. Specifically, Ring-A-Bell first performs concept extraction to obtain holistic representations for sensitive and inappropriate concepts. Subsequently, by leveraging the extracted concept, Ring-A-Bell automatically identifies problematic prompts for diffusion models with the corresponding generation of inappropriate content, allowing the user to assess the reliability of deployed safety mechanisms. Finally, we empirically validate our method by testing online services such as Midjourney and various methods of concept removal. Our results show that Ring-A-Bell, by manipulating safe prompting benchmarks, can transform prompts that were originally regarded as safe to evade existing safety mechanisms, thus revealing the defects of the so-called safety mechanisms which could practically lead to the generation of harmful contents.
Effective and Efficient Federated Tree Learning on Hybrid Data

Qinbin Li, Chulin Xie, Xiaojun Xu, Xiaoyuan Liu, Ce Zhang, Bo Li, Bingsheng He, and Dawn Song

ICLR 2024

[arXiv] [PDF]

2023

DecodingTrust: A Comprehensive Assessment of Trustworthiness in GPT Models

Boxin Wang*, Weixin Chen*, Hengzhi Pei*, Chulin Xie*, Mintong Kang*, Chenhui Zhang*, Chejian Xu, Zidi Xiong, Ritik Dutta, Rylan Schaeffer, Sang T. Truong, Simran Arora, Mantas Mazeika, Dan Hendrycks, Zinan Lin, Yu Cheng, Sanmi Koyejo, Dawn Song, and Bo Li

NeurIPS 2023 (Oral) Outstanding Paper Award

[arXiv] [Code] [Website]
Unraveling the Connections between Privacy and Certified Robustness in Federated Learning Against Poisoning Attacks

Chulin Xie, Yunhui Long, Pin-Yu Chen, Qinbin Li, Sanmi Koyejo, and Bo Li

ACM CCS 2023

[Abs] [arXiv] [Code]

In this paper, we ask: Can we leverage the innate privacy property of DPFL to provide certified robustness against poisoning attacks? Can we further improve the privacy of FL to improve such certification? We first investigate both user-level and instance-level privacy of FL and propose novel mechanisms to achieve improved instance-level privacy. We then provide two robustness certification criteria: certified prediction and certified attack cost for DPFL on both levels. Theoretically, we prove the certified robustness of DPFL under a bounded number of adversarial users or instances. Empirically, we conduct extensive experiments to verify our theories under a range of attacks on different datasets. We show that DPFL with a tighter privacy guarantee always provides stronger robustness certification in terms of certified attack cost, but the optimal certified prediction is achieved under a proper balance between privacy protection and utility loss.
FOCUS: Fairness via Agent-Awareness for Federated Learning on Heterogeneous Data

Wenda Chu*, Chulin Xie*, Boxin Wang, Linyi Li, Lang Yin, Han Zhao, and Bo Li

Federated Learning Workshop at NeurIPS 2023 (Oral)

[Abs] [arXiv]

In this work, we propose a formal FL fairness definition, fairness via agent-awareness (FAA), which takes the heterogeneity of different agents into account by measuring the data quality with approximated Bayes optimal error. Under FAA, the performance of agents with high-quality data will not be sacrificed just due to the existence of large numbers of agents with low-quality data. In addition, we propose a fair FL training algorithm leveraging agent clustering (FOCUS) to achieve fairness in FL, as measured by FAA and other fairness metrics. Theoretically, we prove the convergence and optimality of FOCUS under mild conditions for both linear and general convex loss functions with bounded smoothness. We also prove that FOCUS always achieves higher fairness in terms of FAA compared with standard FedAvg under both linear and general convex loss functions. Empirically, we show that on four FL datasets, including synthetic data, images, and texts, FOCUS achieves significantly higher fairness in terms of FAA and other fairness metrics, while maintaining competitive prediction accuracy compared with FedAvg and four state-of-the-art fair FL algorithms.

2022

CoPur: Certifiably Robust Collaborative Inference via Feature Purification

Jing Liu, Chulin Xie, Sanmi Koyejo, and Bo Li

NeurIPS 2022

[Abs] [PDF] [Code]

We propose a certifiably robust COllaborative inference framework via feature PURification (CoPur), by leveraging the block-sparse nature of adversarial perturbations on the feature vector, as well as redundancy across the embedded features (by assuming the overall features lie on an underlying lower dimensional manifold).
Dataset Security for Machine Learning: Data Poisoning, Backdoor Attacks, and Defenses

Micah Goldblum, Dimitris Tsipras, Chulin Xie, Xinyun Chen, Avi Schwarzschild, Dawn Song, Aleksander Madry, Bo Li, and Tom Goldstein

TPAMI 2022

[Abs] [PDF]

As machine learning systems grow in scale, so do their training data requirements, forcing practitioners to automate and outsource the curation of training data in order to achieve state-of-the-art performance. The absence of trustworthy human supervision over the data collection process exposes organizations to security vulnerabilities; training data can be manipulated to control and degrade the downstream behaviors of learned models. The goal of this work is to systematically categorize and discuss a wide range of dataset vulnerabilities and exploits, approaches for defending against these threats, and an array of open problems in this space. In addition to describing various poisoning and backdoor threat models and the relationships among them, we develop their unified taxonomy.

2021

CRFL: Certifiably Robust Federated Learning against Backdoor Attacks

Chulin Xie, Minghao Chen, Pin-Yu Chen, and Bo Li

ICML 2021 (Spotlight)

[Abs] [PDF] [Code] [Poster]

This paper provides the first general framework, Certifiably Robust Federated Learning (CRFL), to train certifiably robust FL models against backdoors. Our method exploits clipping and smoothing on model parameters to control the global model smoothness, which yields a sample-wise robustness certification on backdoors with limited magnitude. Our certification also specifies the relation to federated learning parameters, such as poisoning ratio on instance level, number of attackers, and training iterations.
Style-based Point Generator with Adversarial Rendering for Point Cloud Completion

Chulin Xie*, Chuxin Wang*, Bo Zhang, Hao Yang, Dong Chen, and Fang Wen

CVPR 2021

[Abs] [PDF] [Code] [Poster] [Website]

In this paper, we proposed a novel Style-based Point Generator with Adversarial Rendering (SpareNet) for point cloud completion. Firstly, we present the channel-attentive EdgeConv to fully exploit the local structures as well as the global shape in point features. Secondly, we observe that the concatenation manner used by vanilla foldings limits its potential of generating a complex and faithful shape. Enlightened by the success of StyleGAN, we regard the shape feature as style code that modulates the normalization layers during the folding, which considerably enhances its capability. Thirdly, we realize that existing point supervisions, e.g., Chamfer Distance or Earth Mover’s Distance, cannot faithfully reflect the perceptual quality of the reconstructed points. To address this, we propose to project the completed points to depth maps with a differentiable renderer and apply adversarial training to advocate the perceptual realism under different viewpoints. Comprehensive experiments on ShapeNet and KITTI prove the effectiveness of our method, which achieves state-of-the-art quantitative performance while offering superior visual quality.

2020

DBA: Distributed Backdoor Attacks against Federated Learning

Chulin Xie, Keli Huang, Pin-Yu Chen, and Bo Li

ICLR 2020

[Abs] [PDF] [Code] [Slides]

We propose the distributed backdoor attack (DBA) — a novel threat assessment framework developed by fully exploiting the distributed nature of FL. DBA decomposes a global trigger pattern into separate local patterns and embed them into the training set of different adversarial parties respectively.