Search Results for author: Mengnan Du

Found 45 papers, 9 papers with code

Mitigating Shortcuts in Language Models with Soft Label Encoding

no code implementations17 Sep 2023 Zirui He, Huiqi Deng, Haiyan Zhao, Ninghao Liu, Mengnan Du

Recent research has shown that large language models rely on spurious correlations in the data for natural language understanding (NLU) tasks.

Natural Language Understanding Out-of-Distribution Generalization

Adaptive Priority Reweighing for Generalizing Fairness Improvement

1 code implementation15 Sep 2023 Zhihao Hu, Yiran Xu, Mengnan Du, Jindong Gu, Xinmei Tian, Fengxiang He

Our adaptive reweighing method prioritizes samples closer to the decision boundary and assigns a higher weight to improve the generalizability of fair classifiers.

Decision Making Fairness

Explainability for Large Language Models: A Survey

no code implementations2 Sep 2023 Haiyan Zhao, Hanjie Chen, Fan Yang, Ninghao Liu, Huiqi Deng, Hengyi Cai, Shuaiqiang Wang, Dawei Yin, Mengnan Du

For each paradigm, we summarize the goals and dominant approaches for generating local explanations of individual predictions and global explanations of overall model knowledge.

A Survey on Fairness in Large Language Models

no code implementations20 Aug 2023 Yingji Li, Mengnan Du, Rui Song, Xin Wang, Ying Wang

Then, for large-scale LLMs, we introduce recent fairness research, including fairness evaluation, reasons for bias, and debiasing methods.


XGBD: Explanation-Guided Graph Backdoor Detection

1 code implementation8 Aug 2023 Zihan Guan, Mengnan Du, Ninghao Liu

An emerging detection strategy in the vision and NLP domains is based on an intriguing phenomenon: when training models on a mixture of backdoor and clean samples, the loss on backdoor samples drops significantly faster than on clean samples, allowing backdoor samples to be easily detected by selecting samples with the lowest loss values.

Graph Learning

DISPEL: Domain Generalization via Domain-Specific Liberating

no code implementations14 Jul 2023 Chia-Yuan Chang, Yu-Neng Chuang, Guanchu Wang, Mengnan Du, Na Zou

Domain generalization aims to learn a generalization model that can perform well on unseen test domains by only training on limited source domains.

Domain Generalization

FAIRER: Fairness as Decision Rationale Alignment

no code implementations27 Jun 2023 Tianlin Li, Qing Guo, Aishan Liu, Mengnan Du, Zhiming Li, Yang Liu

Existing fairness regularization terms fail to achieve decision rationale alignment because they only constrain last-layer outputs while ignoring intermediate neuron alignment.


Black-box Backdoor Defense via Zero-shot Image Purification

no code implementations21 Mar 2023 Yucheng Shi, Mengnan Du, Xuansheng Wu, Zihan Guan, Ninghao Liu

Our proposed framework can be applied to black-box models without requiring any internal information about the poisoned model or any prior knowledge of the clean/poisoned samples.

backdoor defense

Understanding and Unifying Fourteen Attribution Methods with Taylor Interactions

no code implementations2 Mar 2023 Huiqi Deng, Na Zou, Mengnan Du, Weifu Chen, Guocan Feng, Ziwei Yang, Zheyang Li, Quanshi Zhang

Various attribution methods have been developed to explain deep neural networks (DNNs) by inferring the attribution/importance/contribution score of each input variable to the final output.

Efficient XAI Techniques: A Taxonomic Survey

no code implementations7 Feb 2023 Yu-Neng Chuang, Guanchu Wang, Fan Yang, Zirui Liu, Xuanting Cai, Mengnan Du, Xia Hu

Finally, we summarize the challenges of deploying XAI acceleration methods to real-world scenarios, overcoming the trade-off between faithfulness and efficiency, and the selection of different acceleration methods.

Explainable artificial intelligence Explainable Artificial Intelligence (XAI)

Mitigating Relational Bias on Knowledge Graphs

no code implementations26 Nov 2022 Yu-Neng Chuang, Kwei-Herng Lai, Ruixiang Tang, Mengnan Du, Chia-Yuan Chang, Na Zou, Xia Hu

Knowledge graph data are prevalent in real-world applications, and knowledge graph neural networks (KGNNs) are essential techniques for knowledge graph representation learning.

Graph Representation Learning Knowledge Graphs +1

Mitigating Algorithmic Bias with Limited Annotations

1 code implementation20 Jul 2022 Guanchu Wang, Mengnan Du, Ninghao Liu, Na Zou, Xia Hu

Existing work on fairness modeling commonly assumes that sensitive attributes for all instances are fully available, which may not be true in many real-world applications due to the high cost of acquiring sensitive information.


Fair Machine Learning in Healthcare: A Review

no code implementations29 Jun 2022 Qizhang Feng, Mengnan Du, Na Zou, Xia Hu

Benefiting from the digitization of healthcare data and the development of computing power, machine learning methods are increasingly used in the healthcare domain.

BIG-bench Machine Learning Fairness

Accelerating Shapley Explanation via Contributive Cooperator Selection

1 code implementation17 Jun 2022 Guanchu Wang, Yu-Neng Chuang, Mengnan Du, Fan Yang, Quan Zhou, Pushkar Tripathi, Xuanting Cai, Xia Hu

Even though Shapley value provides an effective explanation for a DNN model prediction, the computation relies on the enumeration of all possible input feature coalitions, which leads to the exponentially growing complexity.

Unveiling Project-Specific Bias in Neural Code Models

no code implementations19 Jan 2022 Zhiming Li, Yanzhou Li, Tianlin Li, Mengnan Du, Bozhi Wu, Yushi Cao, Xiaofei Xie, Yi Li, Yang Liu

Neural code models have introduced significant improvements over many software analysis tasks like type inference, vulnerability detection, etc.

Adversarial Robustness Vulnerability Detection

Robustness Challenges in Model Distillation and Pruning for Natural Language Understanding

no code implementations16 Oct 2021 Mengnan Du, Subhabrata Mukherjee, Yu Cheng, Milad Shokouhi, Xia Hu, Ahmed Hassan Awadallah

Recent work has focused on compressing pre-trained language models (PLMs) like BERT where the major focus has been to improve the in-distribution performance for downstream tasks.

Knowledge Distillation Model Compression +1

Was my Model Stolen? Feature Sharing for Robust and Transferable Watermarks

no code implementations29 Sep 2021 Ruixiang Tang, Hongye Jin, Curtis Wigington, Mengnan Du, Rajiv Jain, Xia Hu

The main idea is to insert a watermark which is only known to defender into the protected model and the watermark will then be transferred into all stolen models.

Model extraction

Fairness via Representation Neutralization

no code implementations NeurIPS 2021 Mengnan Du, Subhabrata Mukherjee, Guanchu Wang, Ruixiang Tang, Ahmed Hassan Awadallah, Xia Hu

This process not only requires a lot of instance-level annotations for sensitive attributes, it also does not guarantee that all fairness sensitive information has been removed from the encoder.

Classification Fairness

A General Taylor Framework for Unifying and Revisiting Attribution Methods

no code implementations28 May 2021 Huiqi Deng, Na Zou, Mengnan Du, Weifu Chen, Guocan Feng, Xia Hu

However, the attribution problem has not been well-defined, which lacks a unified guideline to the contribution assignment process.

Benchmarking Decision Making

Learning Disentangled Representations for Time Series

no code implementations17 May 2021 Yuening Li, Zhengzhang Chen, Daochen Zha, Mengnan Du, Denghui Zhang, Haifeng Chen, Xia Hu

Motivated by the success of disentangled representation learning in computer vision, we study the possibility of learning semantic-rich time-series representations, which remains unexplored due to three main challenges: 1) sequential data structure introduces complex temporal correlations and makes the latent representations hard to interpret, 2) sequential models suffer from KL vanishing problem, and 3) interpretable semantic concepts for time-series often rely on multiple factors instead of individuals.

Disentanglement Time Series +1

Mutual Information Preserving Back-propagation: Learn to Invert for Faithful Attribution

no code implementations14 Apr 2021 Huiqi Deng, Na Zou, Weifu Chen, Guocan Feng, Mengnan Du, Xia Hu

The basic idea is to learn a source signal by back-propagation such that the mutual information between input and output should be as much as possible preserved in the mutual information between input and the source signal.

Decision Making

Towards Interpreting and Mitigating Shortcut Learning Behavior of NLU Models

no code implementations NAACL 2021 Mengnan Du, Varun Manjunatha, Rajiv Jain, Ruchi Deshpande, Franck Dernoncourt, Jiuxiang Gu, Tong Sun, Xia Hu

These two observations are further employed to formulate a measurement which can quantify the shortcut degree of each training sample.

Generative Counterfactuals for Neural Networks via Attribute-Informed Perturbation

no code implementations18 Jan 2021 Fan Yang, Ninghao Liu, Mengnan Du, Xia Hu

With the wide use of deep neural networks (DNN), model interpretability has become a critical concern, since explainable decisions are preferred in high-stake scenarios.

Deep Serial Number: Computational Watermarking for DNN Intellectual Property Protection

no code implementations17 Nov 2020 Ruixiang Tang, Mengnan Du, Xia Hu

In this paper, we present DSN (Deep Serial Number), a simple yet effective watermarking algorithm designed specifically for deep neural networks (DNNs).

Knowledge Distillation

A Unified Taylor Framework for Revisiting Attribution Methods

no code implementations21 Aug 2020 Huiqi Deng, Na Zou, Mengnan Du, Weifu Chen, Guocan Feng, Xia Hu

Attribution methods have been developed to understand the decision-making process of machine learning models, especially deep neural networks, by assigning importance scores to individual features.

Benchmarking Decision Making

An Embarrassingly Simple Approach for Trojan Attack in Deep Neural Networks

1 code implementation15 Jun 2020 Ruixiang Tang, Mengnan Du, Ninghao Liu, Fan Yang, Xia Hu

In this paper, we investigate a specific security problem called trojan attack, which aims to attack deployed DNN systems relying on the hidden trigger patterns inserted by malicious hackers.

Mitigating Gender Bias in Captioning Systems

1 code implementation15 Jun 2020 Ruixiang Tang, Mengnan Du, Yuening Li, Zirui Liu, Na Zou, Xia Hu

Image captioning has made substantial progress with huge supporting image collections sourced from the web.

Benchmarking Gender Prediction +1

Adversarial Attacks and Defenses: An Interpretation Perspective

no code implementations23 Apr 2020 Ninghao Liu, Mengnan Du, Ruocheng Guo, Huan Liu, Xia Hu

In this paper, we review recent work on adversarial attacks and defenses, particularly from the perspective of machine learning interpretation.

Adversarial Attack Adversarial Defense +2

Score-CAM: Score-Weighted Visual Explanations for Convolutional Neural Networks

9 code implementations3 Oct 2019 Haofan Wang, Zifan Wang, Mengnan Du, Fan Yang, Zijian Zhang, Sirui Ding, Piotr Mardziel, Xia Hu

Recently, increasing attention has been drawn to the internal mechanisms of convolutional neural networks, and the reason why the network makes specific decisions.

Adversarial Attack Decision Making +1

Sub-Architecture Ensemble Pruning in Neural Architecture Search

1 code implementation1 Oct 2019 Yijun Bian, Qingquan Song, Mengnan Du, Jun Yao, Huanhuan Chen, Xia Hu

Neural architecture search (NAS) is gaining more and more attention in recent years due to its flexibility and remarkable capability to reduce the burden of neural network design.

Ensemble Learning Ensemble Pruning +1

Distribution-Guided Local Explanation for Black-Box Classifiers

no code implementations25 Sep 2019 Weijie Fu, Meng Wang, Mengnan Du, Ninghao Liu, Shijie Hao, Xia Hu

Existing local explanation methods provide an explanation for each decision of black-box classifiers, in the form of relevance scores of features according to their contributions.

Towards Generalizable Deepfake Detection with Locality-aware AutoEncoder

no code implementations13 Sep 2019 Mengnan Du, Shiva Pentyala, Yuening Li, Xia Hu

The analysis further shows that LAE outperforms the state-of-the-arts by 6. 52%, 12. 03%, and 3. 08% respectively on three deepfake detection tasks in terms of generalization accuracy on previously unseen manipulations.

Active Learning DeepFake Detection +2

Fairness in Deep Learning: A Computational Perspective

no code implementations23 Aug 2019 Mengnan Du, Fan Yang, Na Zou, Xia Hu

Deep learning is increasingly being used in high-stake decision making applications that affect individual lives.

Decision Making Fairness

Learning Credible Deep Neural Networks with Rationale Regularization

no code implementations13 Aug 2019 Mengnan Du, Ninghao Liu, Fan Yang, Xia Hu

Recent explainability related studies have shown that state-of-the-art DNNs do not always adopt correct evidences to make decisions.

text-classification Text Classification

SpecAE: Spectral AutoEncoder for Anomaly Detection in Attributed Networks

no code implementations11 Aug 2019 Yuening Li, Xiao Huang, Jundong Li, Mengnan Du, Na Zou

SpecAE leverages Laplacian sharpening to amplify the distances between representations of anomalies and the ones of the majority.

Anomaly Detection Density Estimation

Deep Structured Cross-Modal Anomaly Detection

no code implementations11 Aug 2019 Yuening Li, Ninghao Liu, Jundong Li, Mengnan Du, Xia Hu

To this end, we propose a novel deep structured anomaly detection framework to identify the cross-modal anomalies embedded in the data.

Anomaly Detection

Evaluating Explanation Without Ground Truth in Interpretable Machine Learning

no code implementations16 Jul 2019 Fan Yang, Mengnan Du, Xia Hu

Interpretable Machine Learning (IML) has become increasingly important in many real-world applications, such as autonomous cars and medical diagnosis, where explanations are significantly preferred to help people better understand how machine learning systems work and further enhance their trust towards systems.

BIG-bench Machine Learning Interpretable Machine Learning +1

XFake: Explainable Fake News Detector with Visualizations

no code implementations8 Jul 2019 Fan Yang, Shiva K. Pentyala, Sina Mohseni, Mengnan Du, Hao Yuan, Rhema Linder, Eric D. Ragan, Shuiwang Ji, Xia Hu

In this demo paper, we present the XFake system, an explainable fake news detector that assists end-users to identify news credibility.

On Attribution of Recurrent Neural Network Predictions via Additive Decomposition

no code implementations27 Mar 2019 Mengnan Du, Ninghao Liu, Fan Yang, Shuiwang Ji, Xia Hu

REAT decomposes the final prediction of a RNN into additive contribution of each word in the input text.

Decision Making

Techniques for Interpretable Machine Learning

no code implementations31 Jul 2018 Mengnan Du, Ninghao Liu, Xia Hu

Interpretable machine learning tackles the important problem that humans cannot understand the behaviors of complex machine learning models and how these models arrive at a particular decision.

BIG-bench Machine Learning Interpretable Machine Learning

Towards Explanation of DNN-based Prediction with Guided Feature Inversion

no code implementations19 Mar 2018 Mengnan Du, Ninghao Liu, Qingquan Song, Xia Hu

While deep neural networks (DNN) have become an effective computational tool, the prediction results are often criticized by the lack of interpretability, which is essential in many real-world applications such as health informatics.

Decision Making

Cannot find the paper you are looking for? You can Submit a new open access paper.