As LLMs develop increasingly advanced capabilities, there is an increased need to minimize the harm that could be caused to society by certain model outputs; hence, most LLMs have safety guardrails added, for example via fine-tuning. In this paper, we argue the position that current safety fine-tuning is very similar to a traditional cat-and-mouse game (or arms race) between attackers and defenders in cybersecurity. Model jailbreaks and attacks are patched with bandaids to target the specific attack mechanism, but many similar attack vectors might remain. When defenders are not proactively coming up with principled mechanisms, it becomes very easy for attackers to sidestep any new defenses. We show how current defenses are insufficient to prevent new adversarial jailbreak attacks, reward hacking, and loss of control problems. In order to learn from past mistakes in cybersecurity, we draw analogies with historical examples and develop lessons learned that can be applied to LLM safety. These arguments support the need for new and more principled approaches to designing safe models, which are architected for security from the beginning. We describe several such approaches from the AI literature.

In this paper, we experimentally analyze the robustness of selected Federated Learning (FL) systems in the presence of adversarial clients. We find that temporal attacks significantly affect model performance in the FL models tested, especially when the adversaries are active throughout or during the later rounds. We consider a variety of classic learning models, including Multinominal Logistic Regression (MLR), Random Forest, XGBoost, Support Vector Classifier (SVC), as well as various Neural Network models including Multilayer Perceptron (MLP), Convolution Neural Network (CNN), Recurrent Neural Network (RNN), and Long Short-Term Memory (LSTM). Our results highlight the effectiveness of temporal attacks and the need to develop strategies to make the FL process more robust against such attacks. We also briefly consider the effectiveness of defense mechanisms, including outlier detection in the aggregation algorithm.

The rapid development of artificial intelligence has brought about substantial advancements in the field. One promising direction is the integration of Large Language Models (LLMs) with structured knowledge-based systems. This approach aims to enhance AI capabilities by combining the generative language understanding of LLMs with the precise knowledge representation of structured systems. This survey explores the synergy between LLMs and knowledge bases, focusing on real-world applications and addressing associated technical, operational, and ethical challenges. Through a comprehensive literature review, the study identifies critical issues and evaluates existing solutions. The paper highlights the benefits of integrating generative AI with knowledge bases, including improved data contextualization, enhanced model accuracy, and better utilization of knowledge resources. The findings provide a detailed overview of the current state of research, identify key gaps, and offer actionable recommendations. These insights contribute to advancing AI technologies and support their practical deployment across various sectors.

Effective and reliable control over large language model (LLM) behavior is a significant challenge. While activation steering methods, which add steering vectors to a model's hidden states, are a promising approach, existing techniques often lack precision and interpretability in how they influence model outputs. We introduce Feature Guided Activation Additions (FGAA), a novel activation steering method that leverages insights from Contrastive Activation Addition (CAA) and Sparse Autoencoder-Targeted Steering (SAE-TS). By operating in the latent space of a Sparse Autoencoder (SAE) and employing optimization techniques to select desired SAE features, FGAA constructs precise steering vectors that provide better steering effects while maintaining coherence of steered model outputs. In this regard, evaluations on Gemma-2-2B and Gemma-2-9B models across various steering tasks demonstrate that FGAA outperforms existing steering methods of CAA, SAE decoder steering, and SAE-TS. Our results also highlight important trade-offs between steering scale and general model capabilities that are consistent across all tested steering methods.

Traditional Federated Learning (FL) methods encounter significant challenges when dealing with heterogeneous data and providing personalized solutions for non-IID scenarios. Personalized Federated Learning (PFL) approaches aim to address these issues by balancing generalization and personalization, often through parameter decoupling or partial models that freeze some neural network layers for personalization while aggregating other layers globally. However, existing methods still face challenges of global-local model discrepancy, client drift, and catastrophic forgetting, which degrade model accuracy. To overcome these limitations, we propose pMixFed, a dynamic, layer-wise PFL approach that integrates mixup between shared global and personalized local models. Our method introduces an adaptive strategy for partitioning between personalized and shared layers, a gradual transition of personalization degree to enhance local client adaptation, improved generalization across clients, and a novel aggregation mechanism to mitigate catastrophic forgetting. Extensive experiments demonstrate that pMixFed outperforms state-of-the-art PFL methods, showing faster model training, increased robustness, and improved handling of data heterogeneity under different heterogeneous settings.

Deep neural networks (DNNs) have been shown to be vulnerable to adversarial examples, which can produce erroneous predictions by injecting imperceptible perturbations. In this work, we study the transferability of adversarial examples, which is significant due to its threat to real-world applications where model architecture or parameters are usually unknown. Many existing works reveal that the adversarial examples are likely to overfit the surrogate model that they are generated from, limiting its transfer attack performance against different target models. To mitigate the overfitting of the surrogate model, we propose a novel attack method, dubbed reverse adversarial perturbation (RAP). Specifically, instead of minimizing the loss of a single adversarial point, we advocate seeking adversarial example located at a region with unified low loss value, by injecting the worst-case perturbation (the reverse adversarial perturbation) for each step of the optimization procedure.

In few-shot domain adaptation (FDA), classifiers for the target domain are trained with \emph{accessible} labeled data in the source domain (SD) and few labeled data in the target domain (TD). However, data usually contain private information in the current era, e.g., data distributed on personal phones. Thus, the private data will be leaked if we directly access data in SD to train a target-domain classifier (required by FDA methods). In this paper, to prevent privacy leakage in SD, we consider a very challenging problem setting, where the classifier for the TD has to be trained using few labeled target data and a well-trained SD classifier, named few-shot hypothesis adaptation (FHA). In FHA, we cannot access data in SD, as a result, the private information in SD will be protected well.

Large, diachronic datasets of political discourse are hard to come across, especially for resource-lean languages such as Greek. In this paper, we introduce a curated dataset of the Greek Parliament Proceedings that extends chronologically from 1989 up to 2020. It consists of more than 1 million speeches with extensive meta-data, extracted from 5,355 parliamentary sitting record files. We explain how it was constructed and the challenges that had to be overcome. The dataset can be used for both computational linguistics and political analysis---ideally, combining the two.

FILE - Chinese President Xi Jinping waves at an event to introduce new members of the Politburo Standing Committee at the Great Hall of the People in Beijing on Oct. 23, 2022. A man is seen using the OpenAI ChatGPT artificial intelligence chat website in this illustration photo on July 18, 2023. AMERICA MUST WIN: OpenAI's Chris Lehane is warning of America's shrinking lead in the artificial intelligence space as the company releases its economic blueprint and policy proposals for the U.S. 'ONCE UPON A TIME': A happily-ever-after with someone a woman believed was Hollywood hunk Brad Pitt quickly turned into a living nightmare. AI TRANSFORMER HOMES: AC Future, a leading developer of AI-enabled sustainable living solutions, has partnered with world-renowned Italian design house Pininfarina to create a groundbreaking collection of transformable living spaces. This innovative collaboration has resulted in three distinct products: AI-THd (AI Transformer Home Drivable), AI-THu (AI Transformer Home Unit) and AI-THt (AI Transformer Home Trailer).

Evaluating adversarial robustness amounts to finding the minimum perturbation needed to have an input sample misclassified. The inherent complexity of the underlying optimization requires current gradient-based attacks to be carefully tuned, initialized, and possibly executed for many computationally-demanding iterations, even if specialized to a given perturbation model.In this work, we overcome these limitations by proposing a fast minimum-norm (FMN) attack that works with different \ell_p -norm perturbation models ( p 0, 1, 2, \infty), is robust to hyperparameter choices, does not require adversarial starting points, and converges within few lightweight steps. It works by iteratively finding the sample misclassified with maximum confidence within an \ell_p -norm constraint of size \epsilon, while adapting \epsilon to minimize the distance of the current sample to the decision boundary.Extensive experiments show that FMN significantly outperforms existing \ell_0, \ell_1, and \ell_\infty -norm attacks in terms of perturbation size, convergence speed and computation time, while reporting comparable performances with state-of-the-art \ell_2 -norm attacks.