Search Results for author: Junfeng Yang

Found 34 papers, 21 papers with code

NEUZZ: Efficient Fuzzing with Neural Program Smoothing

1 code implementation15 Jul 2018 Dongdong She, Kexin Pei, Dave Epstein, Junfeng Yang, Baishakhi Ray, Suman Jana

However, even state-of-the-art fuzzers are not very efficient at finding hard-to-trigger software bugs.

Evolutionary Algorithms

DeepXplore: Automated Whitebox Testing of Deep Learning Systems

3 code implementations18 May 2017 Kexin Pei, Yinzhi Cao, Junfeng Yang, Suman Jana

First, we introduce neuron coverage for systematically measuring the parts of a DL system exercised by test inputs.

Malware Detection Self-Driving Cars

Trex: Learning Execution Semantics from Micro-Traces for Binary Similarity

1 code implementation16 Dec 2020 Kexin Pei, Zhou Xuan, Junfeng Yang, Suman Jana, Baishakhi Ray

We thus train the model to learn execution semantics from the functions' micro-traces, without any manual labeling effort.

Transfer Learning Vulnerability Detection

XDA: Accurate, Robust Disassembly with Transfer Learning

1 code implementation2 Oct 2020 Kexin Pei, Jonas Guan, David Williams-King, Junfeng Yang, Suman Jana

We present XDA, a transfer-learning-based disassembly framework that learns different contextual dependencies present in machine code and transfers this knowledge for accurate and robust disassembly.

Language Modelling Masked Language Modeling +2

Formal Security Analysis of Neural Networks using Symbolic Intervals

3 code implementations28 Apr 2018 Shiqi Wang, Kexin Pei, Justin Whitehouse, Junfeng Yang, Suman Jana

In this paper, we present a new direction for formally checking security properties of DNNs without using SMT solvers.

Autonomous Vehicles Collision Avoidance

Efficient Formal Safety Analysis of Neural Networks

2 code implementations NeurIPS 2018 Shiqi Wang, Kexin Pei, Justin Whitehouse, Junfeng Yang, Suman Jana

Our approach can check different safety properties and find concrete counterexamples for networks that are 10$\times$ larger than the ones supported by existing analysis techniques.

Adversarial Attack Adversarial Defense +3

Adversarial Attacks are Reversible with Natural Supervision

1 code implementation ICCV 2021 Chengzhi Mao, Mia Chiquier, Hao Wang, Junfeng Yang, Carl Vondrick

We find that images contain intrinsic structure that enables the reversal of many adversarial attacks.

Doubly Right Object Recognition: A Why Prompt for Visual Rationales

1 code implementation CVPR 2023 Chengzhi Mao, Revant Teotia, Amrutha Sundar, Sachit Menon, Junfeng Yang, Xin Wang, Carl Vondrick

We propose a ``doubly right'' object recognition benchmark, where the metric requires the model to simultaneously produce both the right labels as well as the right rationales.

Object Recognition

Multitask Learning Strengthens Adversarial Robustness

1 code implementation ECCV 2020 Chengzhi Mao, Amogh Gupta, Vikram Nitin, Baishakhi Ray, Shuran Song, Junfeng Yang, Carl Vondrick

Although deep networks achieve strong accuracy on a range of computer vision benchmarks, they remain vulnerable to adversarial attacks, where imperceptible input perturbations fool the network.

Adversarial Defense Adversarial Robustness

Generative Interventions for Causal Learning

1 code implementation CVPR 2021 Chengzhi Mao, Augustine Cha, Amogh Gupta, Hao Wang, Junfeng Yang, Carl Vondrick

We introduce a framework for learning robust visual representations that generalize to new viewpoints, backgrounds, and scene contexts.

Ranked #44 on Image Classification on ObjectNet (using extra training data)

Image Classification Out-of-Distribution Generalization

Live Trojan Attacks on Deep Neural Networks

1 code implementation22 Apr 2020 Robby Costales, Chengzhi Mao, Raphael Norwitz, Bryan Kim, Junfeng Yang

We propose a live attack on deep learning systems that patches model parameters in memory to achieve predefined malicious behavior on a certain set of inputs.

Interpreting and Controlling Vision Foundation Models via Text Explanations

1 code implementation16 Oct 2023 Haozhe Chen, Junfeng Yang, Carl Vondrick, Chengzhi Mao

Large-scale pre-trained vision foundation models, such as CLIP, have become de facto backbones for various vision tasks.

Model Editing Visual Reasoning

Raidar: geneRative AI Detection viA Rewriting

1 code implementation23 Jan 2024 Chengzhi Mao, Carl Vondrick, Hao Wang, Junfeng Yang

We find that large language models (LLMs) are more likely to modify human-written text than AI-generated text when tasked with rewriting.

BPF for storage: an exokernel-inspired approach

1 code implementation25 Feb 2021 Yu Jian Wu, Hongyi Wang, Yuhong Zhong, Asaf Cidon, Ryan Stutsman, Amy Tai, Junfeng Yang

The overhead of the kernel storage path accounts for half of the access latency for new NVMe storage devices.

Operating Systems Databases

Using Multiple Self-Supervised Tasks Improves Model Robustness

1 code implementation7 Apr 2022 Matthew Lawhon, Chengzhi Mao, Junfeng Yang

In this paper, we propose a novel defense that can dynamically adapt the input using the intrinsic structure from multiple self-supervised tasks.

A Tale of Two Models: Constructing Evasive Attacks on Edge Models

1 code implementation22 Apr 2022 Wei Hao, Aahil Awatramani, Jiayang Hu, Chengzhi Mao, Pin-Chun Chen, Eyal Cidon, Asaf Cidon, Junfeng Yang

In this paper, we introduce a new evasive attack, DIVA, that exploits these differences in edge adaptation, by adding adversarial noise to input data that maximizes the output difference between the original and adapted model.

Quantization Vocal Bursts Valence Prediction

Towards Practical Verification of Machine Learning: The Case of Computer Vision Systems

no code implementations5 Dec 2017 Kexin Pei, Linjie Zhu, Yinzhi Cao, Junfeng Yang, Carl Vondrick, Suman Jana

Finally, we show that retraining using the safety violations detected by VeriVis can reduce the average number of violations up to 60. 2%.

BIG-bench Machine Learning Medical Diagnosis

PanNet: A Deep Network Architecture for Pan-Sharpening

no code implementations ICCV 2017 Junfeng Yang, Xueyang Fu, Yuwen Hu, Yue Huang, Xinghao Ding, John Paisley

We incorporate domain-specific knowledge to design our PanNet architecture by focusing on the two aims of the pan-sharpening problem: spectral and spatial preservation.

AdvSPADE: Realistic Unrestricted Attacks for Semantic Segmentation

no code implementations6 Oct 2019 Guangyu Shen, Chengzhi Mao, Junfeng Yang, Baishakhi Ray

Due to the inherent robustness of segmentation models, traditional norm-bounded attack methods show limited effect on such type of models.

Adversarial Attack Segmentation +1

What Does CNN Shift Invariance Look Like? A Visualization Study

1 code implementation9 Nov 2020 Jake Lee, Junfeng Yang, Zhangyang Wang

We present the results of three experiments comparing representations of millions of images with exhaustively shifted objects, examining both local invariance (within a few pixels) and global invariance (across the image frame).

Robust Perception through Equivariance

1 code implementation12 Dec 2022 Chengzhi Mao, Lingyu Zhang, Abhishek Joshi, Junfeng Yang, Hao Wang, Carl Vondrick

In this paper, we introduce a framework that uses the dense intrinsic constraints in natural images to robustify inference.

Adversarial Robustness Instance Segmentation +2

Adversarially Robust Video Perception by Seeing Motion

no code implementations13 Dec 2022 Lingyu Zhang, Chengzhi Mao, Junfeng Yang, Carl Vondrick

Even under adaptive attacks where the adversary knows our defense, our algorithm is still effective.

Adversarial Robustness

Packing Privacy Budget Efficiently

no code implementations26 Dec 2022 Pierre Tholoniat, Kelly Kostopoulou, Mosharaf Chowdhury, Asaf Cidon, Roxana Geambasu, Mathias Lécuyer, Junfeng Yang

This DP budget can be regarded as a new type of compute resource in workloads of multiple ML models training on user data.

Fairness Scheduling

Test-time Detection and Repair of Adversarial Samples via Masked Autoencoder

no code implementations22 Mar 2023 Yun-Yun Tsai, Ju-Chin Chao, Albert Wen, Zhaoyuan Yang, Chengzhi Mao, Tapan Shah, Junfeng Yang

Test-time defenses solve these issues but most existing test-time defenses require adapting the model weights, therefore they do not work on frozen models and complicate model memory management.

Contrastive Learning Management

Monitoring and Adapting ML Models on Mobile Devices

no code implementations12 May 2023 Wei Hao, Zixi Wang, Lauren Hong, Lingxiao Li, Nader Karayanni, Chengzhi Mao, Junfeng Yang, Asaf Cidon

ML models are increasingly being pushed to mobile devices, for low-latency inference and offline operation.

Exploiting Code Symmetries for Learning Program Semantics

no code implementations7 Aug 2023 Kexin Pei, Weichen Li, Qirui Jin, Shuyang Liu, Scott Geng, Lorenzo Cavallaro, Junfeng Yang, Suman Jana

This paper tackles the challenge of teaching code semantics to Large Language Models (LLMs) for program analysis by incorporating code symmetries into the model architecture.

A Single-Loop Algorithm for Decentralized Bilevel Optimization

no code implementations15 Nov 2023 Youran Dong, Shiqian Ma, Junfeng Yang, Chao Yin

Bilevel optimization has received more and more attention recently due to its wide applications in machine learning.

Bilevel Optimization Hyperparameter Optimization

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