The evolution of Artificial Intelligence (AI) has been significantly accelerated by advancements in Large Language Models (LLMs) and Large Multimodal Models (LMMs), gradually showcasing potential cognitive reasoning abilities in problem-solving and scientific discovery (i.e., AI4Science) once exclusive to human intellect. To comprehensively evaluate current models' performance in cognitive reasoning abilities, we introduce OlympicArena, which includes 11,163 bilingual problems across both text-only and interleaved text-image modalities. These challenges encompass a wide range of disciplines spanning seven fields and 62 international Olympic competitions, rigorously examined for data leakage. We argue that the challenges in Olympic competition problems are ideal for evaluating AI's cognitive reasoning due to their complexity and interdisciplinary nature, which are essential for tackling complex scientific challenges and facilitating discoveries. Beyond evaluating performance across various disciplines using answer-only criteria, we conduct detailed experiments and analyses from multiple perspectives. We delve into the models' cognitive reasoning abilities, their performance across different modalities, and their outcomes in process-level evaluations, which are vital for tasks requiring complex reasoning with lengthy solutions. Our extensive evaluations reveal that even advanced models like GPT-4o only achieve a 39.97\% overall accuracy (28.67\% for mathematics and 29.71\% for physics), illustrating current AI limitations in complex reasoning and multimodal integration. Through the OlympicArena, we aim to advance AI towards superintelligence, equipping it to address more complex challenges in science and beyond. We also provide a comprehensive set of resources to support AI research, including a benchmark dataset, an open-source annotation platform, a detailed evaluation tool, and a leaderboard with automatic submission features.

Herein, a finite element scheme with total-field formulation for 3-D frequency-domain marine controlled-source electromagnetic (CSEM) modeling with bathymetry is presented. We employ a finite-element method based on an unstructured tetrahedral mesh to directly solve the total electromagnetic field equation in terms of vector-scalar potentials. In addition, an mesh refinement approach is proposed to accurately simulate marine CSEM responses in cases involving different frequencies, transmitters, and receivers. First, the mesh is locally refined according to the volume constraints as well as the transmitter and receiver positions. Next, the mesh is adaptively refined according to a posteriori error estimation of the finite-element solution. The a posteriori error is calculated based on the gradient recovery technology and weighted using the dual-error weighting (DEW) approach. The scheme is parallelized to improve its computational efficiency for accommodating multiple field sources, receivers, and frequencies. Various numerical experiments demonstrate that the proposed scheme can effectively provide highly accurate solutions for marine CSEM responses, and the scheme is applied to the modeling of hydrocarbon reservoirs with bathymetry and multisources. The effects of bathymetry on marine CSEM responses for both inline and broadside geometries are discussed lastly.