Model Inversion (MI) attacks aim to recover the private training data from the target model, which has raised security concerns about the deployment of DNNs in practice.
In this paper, we conduct a comprehensive study of over 10 fast adversarial training methods in terms of adversarial robustness and training costs.
To give a more comprehensive robustness assessment, we introduce COCO-O(ut-of-distribution), a test dataset based on COCO with 6 types of natural distribution shifts.
Developing a practically-robust automatic speech recognition (ASR) is challenging since the model should not only maintain the original performance on clean samples, but also achieve consistent efficacy under small volume perturbations and large domain shifts.
In this paper, we present CValues, the first Chinese human values evaluation benchmark to measure the alignment ability of LLMs in terms of both safety and responsibility criteria.
Given the testing report, by adopting a simple re-ranking mitigation strategy on these identified disadvantaged groups, we show that the fairness of DRSs can be significantly improved.
We also evaluate some robust models including both adversarially trained models and other robust trained models and find that some models show worse robustness against attribute changes than vanilla models.
We use an encoder to map a facial image and its identity message to a cross-model adversarial example which can disrupt multiple facial manipulation systems to achieve initiative protection.
1 code implementation • 23 Jan 2023 • Avirup Sil, Jaydeep Sen, Bhavani Iyer, Martin Franz, Kshitij Fadnis, Mihaela Bornea, Sara Rosenthal, Scott McCarley, Rong Zhang, Vishwajeet Kumar, Yulong Li, Md Arafat Sultan, Riyaz Bhat, Radu Florian, Salim Roukos
The field of Question Answering (QA) has made remarkable progress in recent years, thanks to the advent of large pre-trained language models, newer realistic benchmark datasets with leaderboards, and novel algorithms for key components such as retrievers and readers.
Additionally, our experiments demonstrate that model selection is non-trivial for OOD detection and should be considered as an integral of the proposed method, which differs from the claim in existing works that proposed methods are universal across different models.
Contrastive Language-Image Pre-trained (CLIP) models have zero-shot ability of classifying an image belonging to "[CLASS]" by using similarity between the image and the prompt sentence "a [CONTEXT] of [CLASS]".
Ranked #1 on Domain Generalization on VLCS
Despite of the superb performance on a wide range of tasks, pre-trained language models (e. g., BERT) have been proved vulnerable to adversarial texts.
Out-of-distribution (OOD) detection is a critical task for ensuring the reliability and safety of deep neural networks in real-world scenarios.
For borrowing the advantage from NLP-style AT, we propose Discrete Adversarial Training (DAT).
Ranked #1 on Domain Generalization on Stylized-ImageNet
For the class-center involved triplet loss, the positive and negative samples in each triplet are replaced by their corresponding class centers, which enforces data representations of the same class closer to the class center.
However, the spread impact of emotions in a conversation is rarely addressed in existing researches.
Ranked #7 on Emotion Recognition in Conversation on MELD
Neural passage retrieval is a new and promising approach in open retrieval question answering.
Deep learning models exhibit a preference for statistical fitting over logical reasoning.
In this paper, we propose a Graph Pattern Loss based Diversified Attention Network(GPLDAN) for unsupervised cross-modal retrieval to deeply analyze correlations among representations.
Extensive experiments on two real-world tasks show that AdvGraph exhibits better performance compared with previous work: (i) effective - it significantly strengthens the model robustness even under the adaptive attacks setting without negative impact on model performance over legitimate input; (ii) generic - its key component, i. e., the representation of connotative adversarial knowledge is task-agnostic, which can be reused in any Chinese-based NLP models without retraining; and (iii) efficient - it is a light-weight defense with sub-linear computational complexity, which can guarantee the efficiency required in practical scenarios.
To the best of our knowledge, it is the largest dataset for brand detection and recognition with rich annotations.
End-to-end question answering (QA) requires both information retrieval (IR) over a large document collection and machine reading comprehension (MRC) on the retrieved passages.
Answer validation in machine reading comprehension (MRC) consists of verifying an extracted answer against an input context and question pair.
Transfer learning techniques are particularly useful in NLP tasks where a sizable amount of high-quality annotated data is difficult to obtain.
To this end, this paper proposes an end-to-end method to solve the task of Pair-wise Aspect and Opinion Terms Extraction (PAOTE).
We firstly propose to explore the asking emphasis of a question as a key factor in DQI.
Ranked #1 on Community Question Answering on Quora Question Pairs
To deal with shadows, we build up an autonomous driving shadow dataset and design a deep neural network to detect shadows automatically.
2 code implementations • • Vittorio Castelli, Rishav Chakravarti, Saswati Dana, Anthony Ferritto, Radu Florian, Martin Franz, Dinesh Garg, Dinesh Khandelwal, Scott McCarley, Mike McCawley, Mohamed Nasr, Lin Pan, Cezar Pendus, John Pitrelli, Saurabh Pujar, Salim Roukos, Andrzej Sakrajda, Avirup Sil, Rosario Uceda-Sosa, Todd Ward, Rong Zhang
We introduce TechQA, a domain-adaptation question answering dataset for the technical support domain.
With increasing popularity in online learning, a surge of E-learning platforms have emerged to facilitate education opportunities for k-12 (from kindergarten to 12th grade) students and with this, a wealth of information on their learning logs are getting recorded.
However, the video length is usually long, and there are hierarchical relationships between frames across events in the video, the performance of RNN based models are decreased.
Ranked #1 on Video Classification on YouTube-8M
Liver lesion segmentation is a difficult yet critical task for medical image analysis.
1 code implementation • 23 Jan 2019 • Wei Li, Chengwei Pan, Rong Zhang, Jiaping Ren, Yuexin Ma, Jin Fang, Feilong Yan, Qichuan Geng, Xinyu Huang, Huajun Gong, Weiwei Xu, Guoping Wang, Dinesh Manocha, Ruigang Yang
Our augmented approach combines the flexibility in a virtual environment (e. g., vehicle movements) with the richness of the real world to allow effective simulation of anywhere on earth.
Deep neural network models have recently draw lots of attention, as it consistently produce impressive results in many computer vision tasks such as image classification, object detection, etc.
The practice in an elderly-care company shows that the FPQM can reduce the number of attributes by 90. 56% with a prediction accuracy of 98. 39%.