Recent advances in large language models (LLMs) have demonstrated strong performance on simple text classification tasks, frequently under zero-shot settings. However, their efficacy declines when tackling complex social media challenges such as propaganda detection, hateful meme classification, and toxicity identification. Much of the existing work has focused on using LLMs to generate synthetic training data, overlooking the potential of LLM-based text preprocessing and semantic augmentation. In this paper, we introduce an approach that prompts LLMs to clean noisy text and provide context-rich explanations, thereby enhancing training sets without substantial increases in data volume. We systematically evaluate on the SemEval 2024 multi-label Persuasive Meme dataset and further validate on the Google Jigsaw toxic comments and Facebook hateful memes datasets to assess generalizability. Our results reveal that zero-shot LLM classification underperforms on these high-context tasks compared to supervised models. In contrast, integrating LLM-based semantic augmentation yields performance on par with approaches that rely on human-annotated data, at a fraction of the cost. These findings underscore the importance of strategically incorporating LLMs into machine learning (ML) pipeline for social media classification tasks, offering broad implications for combating harmful content online.

Life is about to change on the remote island nation of Tuvalu. To great fanfare, Tuvalu – an entirely cash-based society – has unveiled its first ever ATM, marking its move towards financial modernisation. But while the 10,000 people living in that country may be celebrating no longer having to queue at the bank, I fear their happiness will be short-lived. The world's first ATM was introduced in Britain in 1967, but for me the tyranny of machines that promise convenience but erode human contact really began about 20 years ago, in the form of self-checkouts in our local Sainsbury's. Having watched the Terminator movie franchise during my formative years, I railed prophetically against them, aware that it was just a small slippery slope from "unexpected item in the bagging area" to the extinction of the human race.

The rapid evolution of Internet of Things (IoT) technology has led to a significant increase in the number of IoT devices, applications, and services. This surge in IoT devices, along with their widespread presence, has made them a prime target for various cyber-attacks, particularly through IoT botnets. As a result, security has become a major concern within the IoT ecosystem. This study focuses on investigating how the latent dimension impacts the performance of different deep learning classifiers when trained on latent vector representations of the train dataset. The primary objective is to compare the outcomes of these models when encoder components from two cutting-edge architectures: the Vision Transformer (ViT) and the Variational Auto-Encoder (VAE) are utilized to project the high dimensional train dataset to the learned low dimensional latent space. The encoder components are employed to project high-dimensional structured .csv IoT botnet traffic datasets to various latent sizes. Evaluated on N-BaIoT and CICIoT2022 datasets, findings reveal that VAE-encoder based dimension reduction outperforms ViT-encoder based dimension reduction for both datasets in terms of four performance metrics including accuracy, precision, recall, and F1-score for all models which can be attributed to absence of spatial patterns in the datasets the ViT model attempts to learn and extract from image instances.

Despite being used in many engineering and scientific areas such as physics and mathematics and often taught in high school, graphical vector addition turns out to be a topic prone to misconceptions in understanding even at university-level physics classes. To improve the learning experience and the resulting understanding of vectors, we propose to investigate how concreteness fading implemented with the use of augmented reality and tangible robots could help learners to build a strong representation of vector addition. We design a gamified learning environment consisting of three concreteness fading stages and conduct an experiment with 30 participants. Our results shows a positive learning gain. We analyze extensively the behavior of the participants to understand the usage of the technological tools -- augmented reality and tangible robots -- during the learning scenario. Finally, we discuss how the combination of these tools shows real advantages in implementing the concreteness fading paradigm. Our work provides empirical insights into how users utilize concrete visualizations conveyed by a haptic-enabled robot and augmented reality in a learning scenario.

Reasoning over Knowledge Graphs (KGs) plays a pivotal role in knowledge graph completion or question answering systems, providing richer and more accurate triples and attributes. As numerical attributes become increasingly essential in characterizing entities and relations in KGs, the ability to reason over these attributes has gained significant importance. Existing graph-based methods such as Graph Neural Networks (GNNs) and Knowledge Graph Embeddings (KGEs), primarily focus on aggregating homogeneous local neighbors and implicitly embedding diverse triples. However, these approaches often fail to fully leverage the potential of logical paths within the graph, limiting their effectiveness in exploiting the reasoning process. To address these limitations, we propose ChainsFormer, a novel chain-based framework designed to support numerical reasoning. Chainsformer not only explicitly constructs logical chains but also expands the reasoning depth to multiple hops. Specially, we introduces Relation-Attribute Chains (RA-Chains), a specialized logic chain, to model sequential reasoning patterns. ChainsFormer captures the step-by-step nature of multi-hop reasoning along RA-Chains by employing sequential in-context learning. To mitigate the impact of noisy chains, we propose a hyperbolic affinity scoring mechanism that selects relevant logic chains in a variable-resolution space. Furthermore, ChainsFormer incorporates an attention-based numerical reasoner to identify critical reasoning paths, enhancing both reasoning accuracy and transparency. Experimental results demonstrate that ChainsFormer significantly outperforms state-of-the-art methods, achieving up to a 20.0% improvement in performance. The implementations are available at https://github.com/zhaodazhuang2333/ChainsFormer.

The NIME conference traditionally focuses on interfaces for music and musical expression. In this paper we reverse this tradition to ask, can interfaces developed for music be successfully appropriated to non-musical applications? To help answer this question we designed and developed a new device, which uses interface metaphors borrowed from analogue synthesisers and audio mixing to physically control the intangible aspects of a Large Language Model. We compared two versions of the device, with and without the audio-inspired augmentations, with a group of artists who used each version over a one week period. Our results show that the use of audio-like controls afforded more immediate, direct and embodied control over the LLM, allowing users to creatively experiment and play with the device over its non-mixer counterpart. Our project demonstrates how cross-sensory metaphors can support creative thinking and embodied practice when designing new technological interfaces.

As our information environments become ever more powered by artificial intelligence (AI), the phenomenon of trust in a human's interactions with this intelligence is becoming increasingly pertinent. For example, in the not too distant future, there will be teams of humans and intelligent robots involved in dealing with the repercussions of high-risk disaster situations such as hurricanes, earthquakes, or nuclear accidents. Even in such conditions of high uncertainty, humans and intelligent machines will need to engage in shared decision making, and trust is fundamental to the effectiveness of these interactions. A key challenge in modeling the dynamics of this trust is to provide a means to incorporate sensitivity to fluctuations in human trust judgments. In this article, we explore the ability of Quantum Random Walk models to model the dynamics of trust in human-AI interactions, and to integrate a sensitivity to fluctuations in participant trust judgments based on the nature of the interaction with the AI. We found that using empirical parameters to inform the use of different Hamiltonians can provide a promising means to model the evolution of trust in Human-AI interactions.

Machine unlearning poses the challenge of ``how to eliminate the influence of specific data from a pretrained model'' in regard to privacy concerns. While prior research on approximated unlearning has demonstrated accuracy and efficiency in time complexity, we claim that it falls short of achieving exact unlearning, and we are the first to focus on fairness and robustness in machine unlearning algorithms. Our study presents fairness Conjectures for a well-trained model, based on the variance-bias trade-off characteristic, and considers their relevance to robustness. Our Conjectures are supported by experiments conducted on the two most widely used model architectures, ResNet and ViT, demonstrating the correlation between fairness and robustness: \textit{the higher fairness-gap is, the more the model is sensitive and vulnerable}. In addition, our experiments demonstrate the vulnerability of current state-of-the-art approximated unlearning algorithms to adversarial attacks, where their unlearned models suffer a significant drop in accuracy compared to the exact-unlearned models. We claim that our fairness-gap measurement and robustness metric should be used to evaluate the unlearning algorithm. Furthermore, we demonstrate that unlearning in the intermediate and last layers is sufficient and cost-effective for time and memory complexity.

With the growing electric vehicles (EVs) charging demand, urban planners face the challenges of providing charging infrastructure at optimal locations. For example, range anxiety during long-distance travel and the inadequate distribution of residential charging stations are the major issues many cities face. To achieve reasonable estimation and deployment of the charging demand, we develop a data-driven system based on existing EV trips in New South Wales (NSW) state, Australia, incorporating multiple factors that enhance the geographical feasibility of recommended charging stations. Our system integrates data sources including EV trip data, geographical data such as route data and Local Government Area (LGA) boundaries, as well as features like fire and flood risks, and Points of Interest (POIs). We visualize our results to intuitively demonstrate the findings from our data-driven, multi-source fusion system, and evaluate them through case studies. The outcome of this work can provide a platform for discussion to develop new insights that could be used to give guidance on where to position future EV charging stations.

Robots can identify recyclable materials by recognizing patterns in colors, textures, shapes and logos. Ever wondered what happens to the recyclables you carefully sort and place in your bin? For years, recycling has been a crucial part of our efforts to reduce waste and protect the environment. However, the recycling industry has faced significant challenges, from rising costs to labor shortages. But what if technology could transform this process, making recycling faster, more efficient and actually effective?