Search Results for author:
Found 16 papers, 3 papers with code
Utilizing Multimodal Feature Consistency to Detect Adversarial Examples on Clinical Summaries
Among the modalities of medical data, the clinical summaries have higher risks to be attacked because they are generated by third-party companies.
Constitutional Classifiers: Defending against Universal Jailbreaks across Thousands of Hours of Red Teaming
Large language models (LLMs) are vulnerable to universal jailbreaks-prompting strategies that systematically bypass model safeguards and enable users to carry out harmful processes that require many model interactions, like manufacturing illegal substances at scale.
Towards Generating Informative Textual Description for Neurons in Language Models
We leverage the potential of generative language models to discover human-interpretable descriptors present in a dataset and use an unsupervised approach to explain neurons with these descriptors.
URET: Universal Robustness Evaluation Toolkit (for Evasion)
In this work, we propose a new framework to enable the generation of adversarial inputs irrespective of the input type and task domain.
Matching Pairs: Attributing Fine-Tuned Models to their Pre-Trained Large Language Models
The wide applicability and adaptability of generative large language models (LLMs) has enabled their rapid adoption.
Robustness of Explanation Methods for NLP Models
Explanation methods have emerged as an important tool to highlight the features responsible for the predictions of neural networks.
Adaptive Verifiable Training Using Pairwise Class Similarity
On CIFAR10, a non-robust LeNet model has a 21. 63% error rate, while a model created using verifiable training and a L-infinity robustness criterion of 8/255, has an error rate of 57. 10%.
Backdoor Smoothing: Demystifying Backdoor Attacks on Deep Neural Networks
Backdoor attacks mislead machine-learning models to output an attacker-specified class when presented a specific trigger at test time.
Supervising Unsupervised Open Information Extraction Models
We propose a novel supervised open information extraction (Open IE) framework that leverages an ensemble of unsupervised Open IE systems and a small amount of labeled data to improve system performance.
Detecting Backdoor Attacks on Deep Neural Networks by Activation Clustering
While machine learning (ML) models are being increasingly trusted to make decisions in different and varying areas, the safety of systems using such models has become an increasing concern.
Defending Against Machine Learning Model Stealing Attacks Using Deceptive Perturbations
Machine learning models are vulnerable to simple model stealing attacks if the adversary can obtain output labels for chosen inputs.
UNSUPERVISED SENTENCE EMBEDDING USING DOCUMENT STRUCTURE-BASED CONTEXT
We present a new unsupervised method for learning general-purpose sentence embeddings.
Probabilistic Prototype Model for Serendipitous Property Mining
Besides providing the relevant information, amusing users has been an important role of the web.
