We propose a novel framework, Stable Diffusion-based Momentum Integrated Adversarial Examples (SD-MIAE), for generating adversarial examples that can effectively mislead neural network classifiers while maintaining visual imperceptibility and preserving the semantic similarity to the original class label. Our method leverages the text-to-image generation capabilities of the Stable Diffusion model by manipulating token embeddings corresponding to the specified class in its latent space. These token embeddings guide the generation of adversarial images that maintain high visual fidelity. The SD-MIAE framework consists of two phases: (1) an initial adversarial optimization phase that modifies token embeddings to produce misclassified yet natural-looking images and (2) a momentum-based optimization phase that refines the adversarial perturbations. By introducing momentum, our approach stabilizes the optimization of perturbations across iterations, enhancing both the misclassification rate and visual fidelity of the generated adversarial examples. Experimental results demonstrate that SD-MIAE achieves a high misclassification rate of 79%, improving by 35% over the state-of-the-art method while preserving the imperceptibility of adversarial perturbations and the semantic similarity to the original class label, making it a practical method for robust adversarial evaluation.

Automated driving systems face challenges in GPS-denied situations. To address this issue, kinematic dead reckoning is implemented using measurements from the steering angle, steering rate, yaw rate, and wheel speed sensors onboard the vehicle. However, dead reckoning methods suffer from drift. This paper provides an arc-length-based map matching method that uses a digital 2D map of the scenario in order to correct drift in the dead reckoning estimate. The kinematic model's prediction is used to introduce a temporal notion to the spatial information available in the map data. Results show reliable improvement in drift for all GPS-denied scenarios tested in this study. This innovative approach ensures that automated vehicles can maintain continuous and reliable navigation, significantly enhancing their safety and operational reliability in environments where GPS signals are compromised or unavailable.

Chatbots like ChatGPT are used for diverse purposes, ranging from resume writing to entertainment. These real-world applications are different from the institutional uses, such as resume screening or credit scoring, which have been the focus of much of AI research on fairness. Ensuring equitable treatment for all users in these first-person contexts is critical. In this work, we study "first-person fairness," which means fairness toward the chatbot user. This includes providing high-quality responses to all users regardless of their identity or background and avoiding harmful stereotypes. We propose a scalable, privacy-preserving method for evaluating one aspect of first-person fairness across a large, heterogeneous corpus of real-world chatbot interactions. Specifically, we assess potential bias linked to users' names, which can serve as proxies for demographic attributes like gender or race, in chatbot systems such as ChatGPT, which provide mechanisms for storing and using user names. Our method leverages a second language model to privately analyze name-sensitivity in the chatbot's responses. We verify the validity of these annotations through independent human evaluation. Further, we show that post-training interventions, including RL, significantly mitigate harmful stereotypes. Our approach also yields succinct descriptions of response differences across tasks. For instance, in the "writing a story" task, chatbot responses show a tendency to create protagonists whose gender matches the likely gender inferred from the user's name. Moreover, a pattern emerges where users with female-associated names receive responses with friendlier and simpler language slightly more often than users with male-associated names. Finally, we provide the system messages required for external researchers to further investigate ChatGPT's behavior with hypothetical user profiles.

The rise of AI-driven manipulation poses significant risks to societal trust and democratic processes. Yet, studying these effects in real-world settings at scale is ethically and logistically impractical, highlighting a need for simulation tools that can model these dynamics in controlled settings to enable experimentation with possible defenses. We present a simulation environment designed to address this. We elaborate upon the Concordia framework that simulates offline, 'real life' activity by adding online interactions to the simulation through social media with the integration of a Mastodon server. We improve simulation efficiency and information flow, and add a set of measurement tools, particularly longitudinal surveys. We demonstrate the simulator with a tailored example in which we track agents' political positions and show how partisan manipulation of agents can affect election results.

Many machine learning (ML) classifiers are claimed to outperform humans, but they still make mistakes that humans do not. The most notorious examples of such mistakes are adversarial visual metamers. This paper aims to define and investigate the phenomenon of adversarial Doppelgangers (AD), which includes adversarial visual metamers, and to compare the performance and robustness of ML classifiers to human performance. We find that AD are inputs that are close to each other with respect to a perceptual metric defined in this paper. AD are qualitatively different from the usual adversarial examples. The vast majority of classifiers are vulnerable to AD and robustness-accuracy trade-offs may not improve them. Some classification problems may not admit any AD robust classifiers because the underlying classes are ambiguous. We provide criteria that can be used to determine whether a classification problem is well defined or not; describe the structure and attributes of an AD-robust classifier; introduce and explore the notions of conceptual entropy and regions of conceptual ambiguity for classifiers that are vulnerable to AD attacks, along with methods to bound the AD fooling rate of an attack. We define the notion of classifiers that exhibit hypersensitive behavior, that is, classifiers whose only mistakes are adversarial Doppelgangers. Improving the AD robustness of hyper-sensitive classifiers is equivalent to improving accuracy. We identify conditions guaranteeing that all classifiers with sufficiently high accuracy are hyper-sensitive. Our findings are aimed at significant improvements in the reliability and security of machine learning systems.

In the high-stakes domain of search-and-rescue missions, the deployment of Unmanned Aerial Vehicles (UAVs) has become increasingly pivotal. These missions require seamless, real-time communication among diverse roles within response teams, particularly between Remote Operators (ROs) and On-Site Operators (OSOs). Traditionally, ROs and OSOs have relied on radio communication to exchange critical information, such as the geolocation of victims, hazardous areas, and points of interest. However, radio communication lacks information visualization, suffers from noise, and requires mental effort to interpret information, leading to miscommunications and misunderstandings. To address these challenges, this paper presents VizCom-AR, an Augmented Reality system designed to facilitate visual communication between ROs and OSOs and their situational awareness during UAV-driven search-and-rescue missions. Our experiments, focus group sessions with police officers, and field study showed that VizCom-AR enhances spatial awareness of both ROs and OSOs, facilitate geolocation information exchange, and effectively complement existing communication tools in UAV-driven emergency response missions. Overall, VizCom-AR offers a fundamental framework for designing Augmented Reality systems for large scale UAV-driven rescue missions.

In this study, we provide constructive proof that Transformers can recognize and generate hierarchical language efficiently with respect to model size, even without the need for a specific positional encoding. Specifically, we show that causal masking and a starting token enable Transformers to compute positional information and depth within hierarchical structures. We demonstrate that Transformers without positional encoding can generate hierarchical languages. Furthermore, we suggest that explicit positional encoding might have a detrimental effect on generalization with respect to sequence length.

The rapid advancement of Multimodal Large Language Models (MLLMs) has been accompanied by the development of various benchmarks to evaluate their capabilities. However, the true nature of these evaluations and the extent to which they assess multimodal reasoning versus merely leveraging the underlying Large Language Model (LLM) backbone remain unclear. This paper presents a comprehensive investigation into the role of LLM backbones in MLLM evaluation, focusing on two critical aspects: the degree to which current benchmarks truly assess multimodal reasoning and the influence of LLM prior knowledge on performance. Specifically, we introduce a modified evaluation protocol to disentangle the contributions of the LLM backbone from multimodal integration, and an automatic knowledge identification technique for diagnosing whether LLMs equip the necessary knowledge for corresponding multimodal questions. Our study encompasses four diverse MLLM benchmarks and eight state-of-the-art MLLMs. Key findings reveal that some benchmarks allow high performance even without visual inputs and up to 50% of error rates can be attributed to insufficient world knowledge in the LLM backbone, indicating a heavy reliance on language capabilities. To address knowledge deficiencies, we propose a knowledge augmentation pipeline that achieves significant performance gains, with improvements of up to 60% on certain datasets, resulting in a approximately 4x increase in performance. Our work provides crucial insights into the role of the LLM backbone in MLLMs, and highlights the need for more nuanced benchmarking approaches. The rapid development of Large Language Models (LLMs) (Touvron et al., 2023; Bai et al., 2023a), combined with advancements in visual encoders (Radford et al., 2021; Zhai et al., 2023) and modality bridge techniques (Liu et al., 2023a; Dai et al., 2023), has catalyzed the evolution of Multimodal Large Language Models (MLLMs) capable of comprehending diverse multi-modal inputs.

Understanding neural networks is crucial to creating reliable and trustworthy deep learning models. Most contemporary research in interpretability analyzes just one model at a time via causal intervention or activation analysis. Yet despite successes, these methods leave significant gaps in our understanding of the training behaviors of neural networks, how their inner representations emerge, and how we can predictably associate model components with task-specific behaviors. Seeking new insights from work in related fields, here we survey literature in the field of model merging, a field that aims to combine the abilities of various neural networks by merging their parameters and identifying task-specific model components in the process. We analyze the model merging literature through the lens of loss landscape geometry, an approach that enables us to connect observations from empirical studies on interpretability, security, model merging, and loss landscape analysis to phenomena that govern neural network training and the emergence of their inner representations. To systematize knowledge in this area, we present a novel taxonomy of model merging techniques organized by their core algorithmic principles. Additionally, we distill repeated empirical observations from the literature in these fields into characterizations of four major aspects of loss landscape geometry: mode convexity, determinism, directedness, and connectivity. We argue that by improving our understanding of the principles underlying model merging and loss landscape geometry, this work contributes to the goal of ensuring secure and trustworthy machine learning in practice.

Recent research has focused on the medium and long-term impacts of generative AI, posing scientific and societal challenges mainly due to the detection and reliability of machine-generated information, which is projected to form the major content on the Web soon. Prior studies show that LLMs exhibit a lower performance in generation tasks (model collapse) as they undergo a fine-tuning process across multiple generations on their own generated content (self-consuming loop). In this paper, we present a comprehensive simulation framework built upon the chat version of LLama2, focusing particularly on the linguistic aspects of the generated content, which has not been fully examined in existing studies. Our results show that the model produces less lexical rich content across generations, reducing diversity. The lexical richness has been measured using the linguistic measures of entropy and TTR as well as calculating the POSTags frequency. The generated content has also been examined with an $n$-gram analysis, which takes into account the word order, and semantic networks, which consider the relation between different words. These findings suggest that the model collapse occurs not only by decreasing the content diversity but also by distorting the underlying linguistic patterns of the generated text, which both highlight the critical importance of carefully choosing and curating the initial input text, which can alleviate the model collapse problem. Furthermore, we conduct a qualitative analysis of the fine-tuned models of the pipeline to compare their performances on generic NLP tasks to the original model. We find that autophagy transforms the initial model into a more creative, doubtful and confused one, which might provide inaccurate answers and include conspiracy theories in the model responses, spreading false and biased information on the Web.