Recent advancements have positioned AI, and particularly Large Language Models (LLMs), as transformative tools for scientific research, capable of addressing complex tasks that require reasoning, problem-solving, and decision-making. Their exceptional capabilities suggest their potential as scientific research assistants but also highlight the need for holistic, rigorous, and domain-specific evaluation to assess effectiveness in real-world scientific applications. This paper describes a multifaceted methodology for Evaluating AI models as scientific Research Assistants (EAIRA) developed at Argonne National Laboratory. This methodology incorporates four primary classes of evaluations. 1) Multiple Choice Questions to assess factual recall; 2) Open Response to evaluate advanced reasoning and problem-solving skills; 3) Lab-Style Experiments involving detailed analysis of capabilities as research assistants in controlled environments; and 4) Field-Style Experiments to capture researcher-LLM interactions at scale in a wide range of scientific domains and applications. These complementary methods enable a comprehensive analysis of LLM strengths and weaknesses with respect to their scientific knowledge, reasoning abilities, and adaptability. Recognizing the rapid pace of LLM advancements, we designed the methodology to evolve and adapt so as to ensure its continued relevance and applicability. This paper describes the methodology state at the end of February 2025. Although developed within a subset of scientific domains, the methodology is designed to be generalizable to a wide range of scientific domains.

LongRoPE2 is a novel approach that extends the effective context window of pre-trained large language models (LLMs) to the target length, while preserving the performance on the original shorter context window. This is achieved by three contributions: (1) a hypothesis that insufficient training in higher RoPE dimensions contributes to the persistent out-of-distribution (OOD) issues observed in existing methods; (2) an effective RoPE rescaling algorithm that adopts evolutionary search guided by "needle-driven" perplexity to address the insufficient training problem; (3) a mixed context window training approach that fine-tunes model weights to adopt rescaled RoPE for long-context sequences while preserving the short-context performance with the original RoPE. Extensive experiments on LLaMA3-8B and Phi3-mini-3.8B across various benchmarks validate the hypothesis and demonstrate the effectiveness of LongRoPE2. Remarkably, LongRoPE2 extends LLaMA3-8B to achieve a 128K effective context length while retaining over 98.5% of short-context performance, using only 10B tokens -- 80x fewer than Meta's approach, which fails to reach the target effective context length. Code will be available at https://github.com/microsoft/LongRoPE.

Sustainability encompasses three key facets: economic, environmental, and social. However, the nascent discourse that is emerging on sustainable artificial intelligence (AI) has predominantly focused on the environmental sustainability of AI, often neglecting the economic and social aspects. Achieving truly sustainable AI necessitates addressing the tension between its climate awareness and its social sustainability, which hinges on equitable access to AI development resources. The concept of resource awareness advocates for broader access to the infrastructure required to develop AI, fostering equity in AI innovation. Yet, this push for improving accessibility often overlooks the environmental costs of expanding such resource usage. In this position paper, we argue that reconciling climate and resource awareness is essential to realizing the full potential of sustainable AI. We use the framework of base-superstructure to analyze how the material conditions are influencing the current AI discourse. We also introduce the Climate and Resource Aware Machine Learning (CARAML) framework to address this conflict and propose actionable recommendations spanning individual, community, industry, government, and global levels to achieve sustainable AI.

However, accurate predictive models are essential for their operation, especially when dealing with fluctuating environmental conditions and increased demand. Traditional hydrological models often struggle to capture the complexity of such systems. The advent of machine learning (ML) offers new opportunities to enhance predictive capabilities by utilizing large datasets and advanced algorithms (Maity et al., 202 4) . This work aims to develop a machine - learning model that predicts dam volume using features such as water area, physical dam attributes, and other characteristics, including full supply capacity. Multiple models were iteratively built to improve predictive accuracy and performance comparison, each incorporating additional features to refine the outputs . Accurately monitoring reservoir storage is challenging since in - situ data are often unavailable; therefore, remote sensing observations of water extent and height combined with data - driven models are i ncreasingly used for reservoir volume estimation ( Ghosh et al., 2014; Hou et al., 2021) . This study seeks to enhance the precision of dam volume estimates, providing a valuable tool for decision - makers in water management.

Time series forecasting is a long-standing problem in statistics and machine learning. One of the key challenges is processing sequences with long-range dependencies. To that end, a recent line of work applied the short-time Fourier transform (STFT), which partitions the sequence into multiple subsequences and applies a Fourier transform to each separately. We propose the Frequency Information Aggregation (FIA)-Net, which is based on a novel complex-valued MLP architecture that aggregates adjacent window information in the frequency domain. To further increase the receptive field of the FIA-Net, we treat the set of windows as hyper-complex (HC) valued vectors and employ HC algebra to efficiently combine information from all STFT windows altogether. Using the HC-MLP backbone allows for improved handling of sequences with long-term dependence. Furthermore, due to the nature of HC operations, the HC-MLP uses up to three times fewer parameters than the equivalent standard window aggregation method. We evaluate the FIA-Net on various time-series benchmarks and show that the proposed methodologies outperform existing state of the art methods in terms of both accuracy and efficiency. Our code is publicly available on https://anonymous.4open.science/r/research-1803/.

Surface contamination on electrical grid insulators leads to an increase in leakage current until an electrical discharge occurs, which can result in a power system shutdown. To mitigate the possibility of disruptive faults resulting in a power outage, monitoring contamination and leakage current can help predict the progression of faults. Given this need, this paper proposes a hybrid deep learning (DL) model for predicting the increase in leakage current in high-voltage insulators. The hybrid structure considers a multi-criteria optimization using tree-structured Parzen estimation, an input stage filter for signal noise attenuation combined with a large language model (LLM) applied for time series forecasting. The proposed optimized LLM outperforms state-of-the-art DL models with a root-mean-square error equal to 2.24$\times10^{-4}$ for a short-term horizon and 1.21$\times10^{-3}$ for a medium-term horizon.

The rivalry between Sam Altman and Elon Musk is entering its Apprentice era. Both men have the ambition to redefine how the modern world works--and both are jockeying for President Donald Trump's blessing to accelerate their plans. Altman's company, OpenAI, as well as Musk's ventures--which include SpaceX, Tesla, and xAI--all depend to some degree on federal dollars, permits, and regulatory support. The president could influence whether OpenAI or xAI produces the next major AI breakthrough, whether Musk can succeed in sending a human to Mars, and whether Altman's big bet on nuclear energy, and fusion reactors in particular, pans out. Understanding the competition between these two men helps illuminate Trump's particular style of governing--one defined by patronage and dealmaking.

As prominent real-time safety-critical reactive control techniques, Control Barrier Function Quadratic Programs (CBF-QPs) work for control affine systems in general but result in local minima in the generated trajectories and consequently cannot ensure convergence to the goals. Contrarily, Modulation of Dynamical Systems (Mod-DSs), including normal, reference, and on-manifold Mod-DS, achieve obstacle avoidance with few and even no local minima but have trouble optimally minimizing the difference between the constrained and the unconstrained controller outputs, and its applications are limited to fully-actuated systems. We dive into the theoretical foundations of CBF-QP and Mod-DS, proving that despite their distinct origins, normal Mod-DS is a special case of CBF-QP, and reference Mod-DS's solutions are mathematically connected to that of the CBF-QP through one equation. Building on top of the unveiled theoretical connections between CBF-QP and Mod-DS, reference Mod-based CBF-QP and on-manifold Mod-based CBF-QP controllers are proposed to combine the strength of CBF-QP and Mod-DS approaches and realize local-minimum-free reactive obstacle avoidance for control affine systems in general. We validate our methods in both simulated hospital environments and real-world experiments using Ridgeback for fully-actuated systems and Fetch robots for underactuated systems. Mod-based CBF-QPs outperform CBF-QPs as well as the optimally constrained-enforcing Mod-DS approaches we proposed in all experiments.

Accurate prediction of shear strength parameters in Municipal Solid Waste (MSW) remains a critical challenge in geotechnical engineering due to the heterogeneous nature of waste materials and their temporal evolution through degradation processes. This paper presents a novel explainable artificial intelligence (XAI) framework for evaluating cohesion and friction angle across diverse MSW compositional profiles. The proposed model integrates a multi-layer perceptron architecture with SHAP (SHapley Additive exPlanations) analysis to provide transparent insights into how specific waste components influence strength characteristics. Training data encompassed large-scale direct shear tests across various waste compositions and degradation states. The model demonstrated superior predictive accuracy compared to traditional gradient boosting methods, achieving mean absolute percentage errors of 7.42% and 14.96% for friction angle and cohesion predictions, respectively. Through SHAP analysis, the study revealed that fibrous materials and particle size distribution were primary drivers of shear strength variation, with food waste and plastics showing significant but non-linear effects. The model's explainability component successfully quantified these relationships, enabling evidence-based recommendations for waste management practices. This research bridges the gap between advanced machine learning and geotechnical engineering practice, offering a reliable tool for rapid assessment of MSW mechanical properties while maintaining interpretability for engineering decision-making.

Full Waveform Inversion (FWI) reconstructs high-resolution subsurface models via multi-variate optimization but faces challenges with solver selection and data availability. Deep Learning (DL) offers a promising alternative, bridging data-driven and physics-based methods. While FWI in DL has been explored in the time domain, the pseudo-spectral approach remains underutilized, despite its success in classical FWI. This thesis integrates pseudo-spectral FWI into DL, formulating both data-driven and theory-guided approaches using Deep Neural Networks (DNNs) and Recurrent Neural Networks (RNNs). These methods were theoretically derived, tested on synthetic and Marmousi datasets, and compared with deterministic and time-domain approaches. Results show that data-driven pseudo-spectral DNNs outperform classical FWI in deeper and over-thrust regions due to their global approximation capability. Theory-guided RNNs yield greater accuracy, with lower error and better fault identification. While DNNs excel in velocity contrast recovery, RNNs provide superior edge definition and stability in shallow and deep sections. Beyond enhancing FWI performance, this research identifies broader applications of DL-based inversion and outlines future directions for these frameworks.