Related papers: BeamClean: Language Aware Embedding Reconstruction
Vision-language models can encode societal biases and stereotypes, but there are challenges to measuring and mitigating these multimodal harms due to lacking measurement robustness and feature degradation. To address these challenges, we…
Backdoor attacks pose a critical threat by embedding hidden triggers into inputs, causing models to misclassify them into target labels. While extensive research has focused on mitigating these attacks in object recognition models through…
Data privacy has emerged as an important issue as data-driven deep learning has been an essential component of modern machine learning systems. For instance, there could be a potential privacy risk of machine learning systems via the model…
In the large language model (LLM) revolution, embedding is a key component of various systems, such as retrieving knowledge or memories for LLMs or building content moderation filters. As such cases span from English to other natural or…
Semantic sentence embedding models encode natural language sentences into vectors, such that closeness in embedding space indicates closeness in the semantics between the sentences. Bilingual data offers a useful signal for learning such…
In this paper, we propose a deep convolutional neural network-based acoustic word embedding system on code-switching query by example spoken term detection. Different from previous configurations, we combine audio data in two languages for…
The evaluation of cross-lingual semantic search models is often limited to existing datasets from tasks such as information retrieval and semantic textual similarity. We introduce Cross-Lingual Semantic Discrimination (CLSD), a lightweight…
Recently, substantial progress has been made in language modeling by using deep neural networks. However, in practice, large scale neural language models have been shown to be prone to overfitting. In this paper, we present a simple yet…
Text classification has become widely used in various natural language processing applications like sentiment analysis. Current applications often use large transformer-based language models to classify input texts. However, there is a lack…
In natural language processing (NLP) of spoken languages, word embeddings have been shown to be a useful method to encode the meaning of words. Sign languages are visual languages, which require sign embeddings to capture the visual and…
In recent years, text generation tools utilizing Artificial Intelligence (AI) have occasionally been misused across various domains, such as generating student reports or creative writings. This issue prompts plagiarism detection services…
We introduce a simple modification to the embedding layer. The key change is to infuse token embeddings with information about their spelling. Models trained with these embeddings improve not only on spelling, but also across standard…
Embeddings are one of the fundamental building blocks for data analysis tasks. Embeddings are already essential tools for large language models and image analysis, and their use is being extended to many other research domains. The…
Model inversion (MI) attacks allow to reconstruct average per-class representations of a machine learning (ML) model's training data. It has been shown that in scenarios where each class corresponds to a different individual, such as face…
Recent inversion methods have shown that real images can be inverted into StyleGAN's latent space and numerous edits can be achieved on those images thanks to the semantically rich feature representations of well-trained GAN models.…
Large language models (LLMs) represent words through contextual word embeddings encoding different language properties like semantics and syntax. Understanding these properties is crucial, especially for researchers investigating language…
To counter man-at-the-end attacks such as reverse engineering and tampering, software is often protected with techniques that require support modules to be linked into the application. It is well-known, however, that attackers can exploit…
In this paper, we introduce the task of learning unsupervised dialogue embeddings. Trivial approaches such as combining pre-trained word or sentence embeddings and encoding through pre-trained language models (PLMs) have been shown to be…
While vision and multimodal foundation models underpin critical tasks from perception to complex reasoning, they remain highly vulnerable to adversarial attacks. However, traditional adversarial attacks are typically limited to single,…
Backdoor attacks manipulate model predictions by inserting innocuous triggers into training and test data. We focus on more realistic and more challenging clean-label attacks where the adversarial training examples are correctly labeled.…