Category: seminar

Abstract: A fundamental problem in modern vector databases is to process approximate similarity queries in high dimensional space. Vector databases become a central research topic with the increasing popularity of large language models. Troubled by the “curse of dimensionality” issue, it has long been questioned whether it is possible to index high-dimensional data effectively and… Continue reading Approximate Similarity Search in Vector Databases

Abstract: Language models (LMs) are essential in natural language processing and vision-language modeling. However, several challenges arise in pre-training and fine-tuning of LMs. First, when learning through unsupervised pre-training, information that are semantically irrelevant may negatively affect downstream tasks, leading to negative transfer. Second, cross-modal masked language modeling is often used to learn vision-language associations… Continue reading Enhancing Language Models through Improved Pre-Training and Fine-Tuning

Abstract: AI can extend human capabilities but requires addressing challenges in education, jobs, and biases. Taking a responsible approach involves understanding AI’s nature, design choices, societal role, and ethical considerations. Recent AI developments, including foundational models, transformer models, generative models, and large language models (LLMs), raise questions about whether they are changing the paradigm of… Continue reading Beyond the AI hype: Balancing Innovation and Social Responsibility

Abstract: Modern distributed systems also deal with uncertain scenarios where environments, infrastructures, and applications are widely diverse. In the scope of IoT-Edge-Fog-Cloud computing, leveraging these neuroscience-inspired principles and mechanisms could aid in building more flexible solutions able to generalize over different environments. A captivating set of hypotheses from the field of neuroscience suggests that human… Continue reading Distributed Intelligence in the Computing Continuum

Abstract: In multiwinner voting, voters report their preferences over the available alternatives, and the goal is to select a fixed-size subset of alternatives, usually referred to as a committee; this model captures a variety of real-life scenarios, from selecting a representative governing body to deciding which search results should appear on the first page of… Continue reading Fairness in Collective Decision-Making: From Multiwinner Voting to Participatory Budgeting

Abstract：Despite the great success of deep neural networks, the adversarial attack can cheat some well-trained classifiers by small permutations. We propose a specific type of adversarial attack that can cheat classifiers by significant changes. Statistically, the existing adversarial attack increases Type II error and the proposed one aims at Type I error, which are hence… Continue reading Research on adversarial attack and robustness of deep neural networks