This week, we're looking back at our hellish 2024 and trying to figure out where to go from here. We began the year with enormous hype around artificial intelligence, but that's cooled off after seeing how useless many AI features have been. It's also clear that many companies, including Microsoft and Apple, are trying to push half-baked AI concepts onto users. Looking forward, we're expecting a rough few years for the tech industry (not to mention the world as a whole). Listen below or subscribe on your podcast app of choice. If you've got suggestions or topics you'd like covered on the show, be sure to email us or drop a note in the comments!

Congressman Tony Gonzales, R-Texas, discusses the future of drone security in the United States during an appearance on'America Reports.' The Federal Aviation Administration issued more restrictions on drone flights across the Northeast on Friday in response to increased sightings in recent weeks. One day after announcing temporary restrictions on most drone flights in New Jersey, the FAA issued 27 No-Fly Zone notices for "special security reasons" in New York on Friday. The restrictions last through Jan. 18, 2025, and apply to some of the most populated areas in the Empire State, including nearly every NYC borough. The Federal Aviation Administration has issued temporary restrictions on drone flights in 27 areas of New York in response to the influx of reported sightings in recent weeks.

We propose a new simulator, training approach, and policy architecture, collectively called SOUS VIDE, for end-to-end visual drone navigation. Our trained policies exhibit zero-shot sim-to-real transfer with robust real-world performance using only on-board perception and computation. Our simulator, called FiGS, couples a computationally simple drone dynamics model with a high visual fidelity Gaussian Splatting scene reconstruction. FiGS can quickly simulate drone flights producing photorealistic images at up to 130 fps. We use FiGS to collect 100k-300k observation-action pairs from an expert MPC with privileged state and dynamics information, randomized over dynamics parameters and spatial disturbances. We then distill this expert MPC into an end-to-end visuomotor policy with a lightweight neural architecture, called SV-Net. SV-Net processes color image, optical flow and IMU data streams into low-level body rate and thrust commands at 20Hz onboard a drone. Crucially, SV-Net includes a Rapid Motor Adaptation (RMA) module that adapts at runtime to variations in drone dynamics. In a campaign of 105 hardware experiments, we show SOUS VIDE policies to be robust to 30% mass variations, 40 m/s wind gusts, 60% changes in ambient brightness, shifting or removing objects from the scene, and people moving aggressively through the drone's visual field. Code, data, and experiment videos can be found on our project page: https://stanfordmsl.github.io/SousVide/.

Existing deep learning-based object detection models perform well under daytime conditions but face significant challenges at night, primarily because they are predominantly trained on daytime images. Additionally, training with nighttime images presents another challenge: even human annotators struggle to accurately label objects in low-light conditions. This issue is particularly pronounced in transportation applications, such as detecting vehicles and other objects of interest on rural roads at night, where street lighting is often absent, and headlights may introduce undesirable glare. This study addresses these challenges by introducing a novel framework for labeling-free data augmentation, leveraging CARLA-generated synthetic data for day-to-night image style transfer. Specifically, the framework incorporates the Efficient Attention Generative Adversarial Network for realistic day-to-night style transfer and uses CARLA-generated synthetic nighttime images to help the model learn vehicle headlight effects. To evaluate the efficacy of the proposed framework, we fine-tuned the YOLO11 model with an augmented dataset specifically curated for rural nighttime environments, achieving significant improvements in nighttime vehicle detection. This novel approach is simple yet effective, offering a scalable solution to enhance AI-based detection systems in low-visibility environments and extend the applicability of object detection models to broader real-world contexts.

NOAA's Deep-ocean Assessment and Reporting of Tsunamis (DART) data are critical for NASA-JPL's tsunami detection, real-time operations, and oceanographic research. However, these time-series data often contain spikes, steps, and drifts that degrade data quality and obscure essential oceanographic features. To address these anomalies, the work introduces an Iterative Encoding-Decoding Variational Autoencoders (Iterative Encoding-Decoding VAEs) model to improve the quality of DART time series. Unlike traditional filtering and thresholding methods that risk distorting inherent signal characteristics, Iterative Encoding-Decoding VAEs progressively remove anomalies while preserving the data's latent structure. A hybrid thresholding approach further retains genuine oceanographic features near boundaries. Applied to complex DART datasets, this approach yields reconstructions that better maintain key oceanic properties compared to classical statistical techniques, offering improved robustness against spike removal and subtle step changes. The resulting high-quality data supports critical verification and validation efforts for the GRACE-FO mission at NASA-JPL, where accurate surface measurements are essential to modeling Earth's gravitational field and global water dynamics. Ultimately, this data processing method enhances tsunami detection and underpins future climate modeling with improved interpretability and reliability.

In federated learning, federated unlearning is a technique that provides clients with a rollback mechanism that allows them to withdraw their data contribution without training from scratch. However, existing research has not considered scenarios with skewed label distributions. Unfortunately, the unlearning of a client with skewed data usually results in biased models and makes it difficult to deliver high-quality service, complicating the recovery process. This paper proposes a recovery method of federated unlearning with skewed label distributions. Specifically, we first adopt a strategy that incorporates oversampling with deep learning to supplement the skewed class data for clients to perform recovery training, therefore enhancing the completeness of their local datasets. Afterward, a density-based denoising method is applied to remove noise from the generated data, further improving the quality of the remaining clients' datasets. Finally, all the remaining clients leverage the enhanced local datasets and engage in iterative training to effectively restore the performance of the unlearning model. Extensive evaluations on commonly used federated learning datasets with varying degrees of skewness show that our method outperforms baseline methods in restoring the performance of the unlearning model, particularly regarding accuracy on the skewed class.

Data-based methods have gained increasing importance in engineering, especially but not only driven by successes with deep artificial neural networks. Success stories are prevalent, e.g., in areas such as data-driven modeling, control and automation, as well as surrogate modeling for accelerated simulation. Beyond engineering, generative and large-language models are increasingly helping with tasks that, previously, were solely associated with creative human processes. Thus, it seems timely to seek artificial-intelligence-support for engineering design tasks to automate, help with, or accelerate purpose-built designs of engineering systems, e.g., in mechanics and dynamics, where design so far requires a lot of specialized knowledge. However, research-wise, compared to established, predominantly first-principles-based methods, the datasets used for training, validation, and test become an almost inherent part of the overall methodology. Thus, data publishing becomes just as important in (data-driven) engineering science as appropriate descriptions of conventional methodology in publications in the past. This article analyzes the value and challenges of data publishing in mechanics and dynamics, in particular regarding engineering design tasks, showing that the latter raise also challenges and considerations not typical in fields where data-driven methods have been booming originally. Possible ways to deal with these challenges are discussed and a set of examples from across different design problems shows how data publishing can be put into practice. The analysis, discussions, and examples are based on the research experience made in a priority program of the German research foundation focusing on research on artificially intelligent design assistants in mechanics and dynamics.

Large Language Models (LLMs) have demonstrated an alarming ability to impersonate humans in conversation, raising concerns about their potential misuse in scams and deception. Humans have a right to know if they are conversing to an LLM. We evaluate text-based prompts designed as challenges to expose LLM imposters in real-time. To this end we compile and release an open-source benchmark dataset that includes 'implicit challenges' that exploit an LLM's instruction-following mechanism to cause role deviation, and 'exlicit challenges' that test an LLM's ability to perform simple tasks typically easy for humans but difficult for LLMs. Our evaluation of 9 leading models from the LMSYS leaderboard revealed that explicit challenges successfully detected LLMs in 78.4% of cases, while implicit challenges were effective in 22.9% of instances. User studies validate the real-world applicability of our methods, with humans outperforming LLMs on explicit challenges (78% vs 22% success rate). Our framework unexpectedly revealed that many study participants were using LLMs to complete tasks, demonstrating its effectiveness in detecting both AI impostors and human misuse of AI tools. This work addresses the critical need for reliable, real-time LLM detection methods in high-stakes conversations.

With the widespread deployment of Multimodal Large Language Models (MLLMs) for visual-reasoning tasks, improving their safety has become crucial. Recent research indicates that despite training-time safety alignment, these models remain vulnerable to jailbreak attacks. In this work, we first highlight an important safety gap to describe that alignment achieved solely through safety training may be insufficient against jailbreak attacks. To address this vulnerability, we propose Immune, an inference-time defense framework that leverages a safe reward model through controlled decoding to defend against jailbreak attacks. Additionally, we provide a mathematical characterization of Immune, offering provable guarantees against jailbreaks. Extensive evaluations on diverse jailbreak benchmarks using recent MLLMs reveal that Immune effectively enhances model safety while preserving the model's original capabilities. For instance, against text-based jailbreak attacks on LLaVA-1.6, Immune reduces the attack success rate by 57.82% and 16.78% compared to the base MLLM and state-of-the-art defense strategy, respectively.

Accelerating model convergence in resource-constrained environments is essential for fast and efficient neural network training. This work presents learn2mix, a new training strategy that adaptively adjusts class proportions within batches, focusing on classes with higher error rates. Empirical evaluations on benchmark datasets show that neural networks trained with learn2mix converge faster than those trained with classical approaches, achieving improved results for classification, regression, and reconstruction tasks under limited training resources and with imbalanced classes. Our empirical findings are supported by theoretical analysis. Despite their ability to learn and model complex, nonlinear relationships, deep neural networks often require substantial computational resources during training. In resource-constrained environments, this demand poses a significant challenge (Goyal et al., 2017), making the development of efficient and scalable training methodologies increasingly crucial to fully leverage the capabilities of deep neural networks. Training deep neural networks relies on the notion of empirical risk minimization (Vapnik & Bottou, 1993), and typically involves optimizing a loss function using gradient-based algorithms (Rumelhart et al., 1986; Bottou, 2010; Kingma & Ba, 2014). Techniques such as regularization (Srivastava et al., 2014; Ioffe & Szegedy, 2015) and data augmentation (Shorten & Khoshgoftaar, 2019), learning rate scheduling, (Smith, 2017) and early stopping (Prechelt, 1998), are commonly employed to enhance generalization and prevent overfitting. However, the efficiency of the training process itself remains a critical concern, particularly in terms of convergence speed and computational resources. Within this context, adaptive training strategies, which target enhanced generalization by modifying aspects of the training process, have emerged as promising approaches.