Effective processing, interpretation, and management of sensor data have emerged as a critical component of cyber-physical systems. Traditionally, processing sensor data requires profound theoretical knowledge and proficiency in signal-processing tools. However, recent works show that Large Language Models (LLMs) have promising capabilities in processing sensory data, suggesting their potential as copilots for developing sensing systems. To explore this potential, we construct a comprehensive benchmark, SensorBench, to establish a quantifiable objective. The benchmark incorporates diverse real-world sensor datasets for various tasks. The results show that while LLMs exhibit considerable proficiency in simpler tasks, they face inherent challenges in processing compositional tasks with parameter selections compared to engineering experts. Additionally, we investigate four prompting strategies for sensor processing and show that self-verification can outperform all other baselines in 48% of tasks. Our study provides a comprehensive benchmark and prompting analysis for future developments, paving the way toward an LLM-based sensor processing copilot.

Laboratory accidents pose significant risks to human life and property, underscoring the importance of robust safety protocols. Despite advancements in safety training, laboratory personnel may still unknowingly engage in unsafe practices. With the increasing reliance on large language models (LLMs) for guidance in various fields, including laboratory settings, there is a growing concern about their reliability in critical safety-related decision-making. Unlike trained human researchers, LLMs lack formal lab safety education, raising questions about their ability to provide safe and accurate guidance. Existing research on LLM trustworthiness primarily focuses on issues such as ethical compliance, truthfulness, and fairness but fails to fully cover safety-critical real-world applications, like lab safety. To address this gap, we propose the Laboratory Safety Benchmark (LabSafety Bench), a comprehensive evaluation framework based on a new taxonomy aligned with Occupational Safety and Health Administration (OSHA) protocols. This benchmark includes 765 multiple-choice questions verified by human experts, assessing LLMs and vision language models (VLMs) performance in lab safety contexts. Our evaluations demonstrate that while GPT-4o outperforms human participants, it is still prone to critical errors, highlighting the risks of relying on LLMs in safety-critical environments. Our findings emphasize the need for specialized benchmarks to accurately assess the trustworthiness of LLMs in real-world safety applications.

Joint Multimodal Entity-Relation Extraction (JMERE) is a challenging task that aims to extract entities and their relations from text-image pairs in social media posts. Existing methods for JMERE require large amounts of labeled data. However, gathering and annotating fine-grained multimodal data for JMERE poses significant challenges. Initially, we construct diverse and comprehensive multimodal few-shot datasets fitted to the original data distribution. To address the insufficient information in the few-shot setting, we introduce the \textbf{K}nowledge-\textbf{E}nhanced \textbf{C}ross-modal \textbf{P}rompt \textbf{M}odel (KECPM) for JMERE. This method can effectively address the problem of insufficient information in the few-shot setting by guiding a large language model to generate supplementary background knowledge. Our proposed method comprises two stages: (1) a knowledge ingestion stage that dynamically formulates prompts based on semantic similarity guide ChatGPT generating relevant knowledge and employs self-reflection to refine the knowledge; (2) a knowledge-enhanced language model stage that merges the auxiliary knowledge with the original input and utilizes a transformer-based model to align with JMERE's required output format. We extensively evaluate our approach on a few-shot dataset derived from the JMERE dataset, demonstrating its superiority over strong baselines in terms of both micro and macro F$_1$ scores. Additionally, we present qualitative analyses and case studies to elucidate the effectiveness of our model.

We revisit the mean field parametrization of shallow neural networks, using signed measures on unbounded parameter spaces and duality pairings that take into account the regularity and growth of activation functions. This setting directly leads to the use of unbalanced Kantorovich-Rubinstein norms defined by duality with Lipschitz functions, and of spaces of measures dual to those of continuous functions with controlled growth. These allow to make transparent the need for total variation and moment bounds or penalization to obtain existence of minimizers of variational formulations, under which we prove a compactness result in strong Kantorovich-Rubinstein norm, and in the absence of which we show several examples demonstrating undesirable behavior. Further, the Kantorovich-Rubinstein setting enables us to combine the advantages of a completely linear parametrization and ensuing reproducing kernel Banach space framework with optimal transport insights. We showcase this synergy with representer theorems and uniform large data limits for empirical risk minimization, and in proposed formulations for distillation and fusion applications.

In processing multiple time series, accounting for the individual features of each sequence can be challenging. To address this, modern deep learning methods for time series analysis combine a shared (global) model with local layers, specific to each time series, often implemented as learnable embeddings. Ideally, these local embeddings should encode meaningful representations of the unique dynamics of each sequence. However, when these are learned end-to-end as parameters of a forecasting model, they may end up acting as mere sequence identifiers. Shared processing blocks may then become reliant on such identifiers, limiting their transferability to new contexts. In this paper, we address this issue by investigating methods to regularize the learning of local learnable embeddings for time series processing. Specifically, we perform the first extensive empirical study on the subject and show how such regularizations consistently improve performance in widely adopted architectures. Furthermore, we show that methods preventing the co-adaptation of local and global parameters are particularly effective in this context. This hypothesis is validated by comparing several methods preventing the downstream models from relying on sequence identifiers, going as far as completely resetting the embeddings during training. The obtained results provide an important contribution to understanding the interplay between learnable local parameters and shared processing layers: a key challenge in modern time series processing models and a step toward developing effective foundation models for time series.

Forecasting of armed conflicts is an important area of research that has the potential to save lives and prevent suffering. However, most existing forecasting models provide only point predictions without any individual-level uncertainty estimates. In this paper, we introduce a novel extension to conformal prediction algorithm which we call bin-conditional conformal prediction. This method allows users to obtain individual-level prediction intervals for any arbitrary prediction model while maintaining a specific level of coverage across user-defined ranges of values. We apply the bin-conditional conformal prediction algorithm to forecast fatalities from armed conflict. Our results demonstrate that the method provides well-calibrated uncertainty estimates for the predicted number of fatalities. Compared to standard conformal prediction, the bin-conditional method outperforms offers improved calibration of coverage rates across different values of the outcome, but at the cost of wider prediction intervals.

Current advances in Natural Language Processing (NLP) have made it increasingly feasible to build applications leveraging textual data. Generally, the core of these applications rely on having a good semantic representation of text into vectors, via embedding models. However, it has been shown that these embeddings capture and perpetuate biases already present in text. While a few techniques have been proposed to debias embeddings, they do not take advantage of the recent advances in context understanding of modern embedding models. In this paper, we fill this gap by conducting a review of 19 embedding models by quantifying their biases and how well they respond to context injection as a mean of debiasing. We show that higher performing models are more prone to capturing biases, but are also better at incorporating context. Surprisingly, we find that while models can easily embed affirmative semantics, they fail at embedding neutral semantics. Finally, in a retrieval task, we show that biases in embeddings can lead to non-desirable outcomes. We use our new-found insights to design a simple algorithm for top $k$ retrieval, where $k$ is dynamically selected. We show that our algorithm is able to retrieve all relevant gendered and neutral chunks.

A humanoid robot is slated to become first of its kind to have its artwork sold by a major auction house. On October 16, Sotheby's announced it will soon begin accepting bids starting at 120,000 for "AI God." The abstract portrait of Alan Turing was painted by Ai-Da, an ongoing, experimental AI-powered robotics project that cites a pivotal 1980's transhumanist feminist manifesto as its inspiration. The auction is scheduled to run from October 31st through November 7th. Completed in 2019 by gallerist Aidan Meller in collaboration with Oxford University researchers and the robotics company, Engineered Arts, Ai-Da uses cameras to capture visual inputs that onboard graphics algorithms then use to formulate generative images with some human guidance and adjustments.

The United States Treasury Department is turning more to artificial intelligence (AI) to fight fraud, using the technology to thwart 4bn in improper payments in the last year. The agency released the estimate in a press release Thursday announcing the success of its "technology and data-driven approach". In fiscal year 2024, which ran from October 2023 to September 2024, the Treasury used machine-learning AI to halt 1bn in cheque fraud, it said. At the same time, its AI processes helped weed out 3bn in other improper payments, including by pinpointing at-risk transactions and improving screening, it added. The 4bn total annual fraud prevention was six times higher than that captured in the previous year, according to the agency.

The United States Department of the Treasury has announced sanctions against Chinese makers of drone engines and parts that President Joe Biden's administration says have directly helped Russia mount long-range attacks in the war in Ukraine. The sanctions, issued on Thursday, target three entities and one individual for their involvement in the development and production of Russia's "Garpiya series" long-range attack drones. "The Garpiya has been deployed by Russia in its brutal war against Ukraine, destroying critical infrastructure and causing mass casualties," the Treasury Department said in a statement announcing the measures. "Designed and developed by People's Republic of China (PRC)-based experts, the Garpiya is produced at PRC-based factories in collaboration with Russian defense firms before transferring the drones to Russia for use against Ukraine." Russia has recently used long-range drone attacks to penetrate Ukraine's air defences, wreaking havoc across the country, including a missile strike in the city of Poltava that killed 55 people and wounded 328.