Search for: All records

This article examines the challenges and opportunities in extracting causal information from text with Large Language Models (LLMs). It first establishes the importance of causality extraction and then explores different views on causality, including common sense ideas informing different data annotation schemes, Aristotle’s Four Causes, and Pearl’s Ladder of Causation. The paper notes the relevance of this conceptual variety for the task. The text reviews datasets and work related to finding causal expressions, both using traditional machine learning methods and LLMs. Although the known limitations of LLMs—hallucinations and lack of common sense—affect the reliability of causal findings, GPT and Gemini models (GPT-5 and Gemini 2.5 Pro and others) show the ability to conduct causality analysis; moreover, they can even apply different perspectives, such as counterfactual and Aristotelian. They are also capable of explaining and critiquing causal analyses: we report an experiment showing that in addition to largely flawless analyses, the newer models exhibit very high agreement of 88–91% on causal relationships between events—much higher than the typically reported inter-annotator agreement of 30–70%. The article concludes with a discussion of the lessons learned about these challenges and questions how LLMs might help address them in the future. For example, LLMs should help address the sparsity of annotated data. Moreover, LLMs point to a future where causality analysis in texts focuses not on annotations but on understanding, as causality is about semantics and not word spans. The Appendices and shared data show examples of LLM outputs on tasks involving causal reasoning and causal information extraction, demonstrating the models’ current abilities and limits. 

University students have begun to use Artificial Intelligence (AI) in many different ways in their undergraduate education, some beneficial to their learning, and some simply expedient to completing assignments with as little work as possible. This exploratory qualitative study examines how undergraduate students used AI in a large General Education course on sustainability and technology at a research university in the United States in 2023. Thirty-nine students documented their use of AI in their final course project, which involved analyzing conceptual networks connecting core sustainability concepts. Through iterative qualitative coding, we identified key patterns in students’ AI use, including higher-order writing tasks (understanding complex topics, finding evidence), lower-order writing tasks (revising, editing, proofreading), and other learning activities (efficiency enhancement, independent research). Students primarily used AI to improve communication of their original ideas, though some leveraged it for more complex tasks like finding evidence and developing arguments. Many students expressed skepticism about AI-generated content and emphasized maintaining their intellectual independence. While some viewed AI as vital for improving their work, others explicitly distinguished between AI-assisted editing and their original thinking. This analysis provides insight into how students navigate AI use when it is explicitly permitted in coursework, with implications for effectively integrating AI into higher education to support student learning. 

Geometric frustration is recognized to generate complex morphologies in self-assembling particulate and molecular systems. In bulk states, frustration drives structured arrays of topological defects. In the dilute limit, these systems have been shown to form a novel state of self-limiting assembly, in which the equilibrium size of multiparticle domains are finite and well defined. In this article we employ Monte Carlo simulations of a recently developed 2D lattice model of geometrically frustrated assembly [Hackney et al., Phys. Rev. X 13, 041010 (2023)] to study the phase transitions between the self-limiting and defect bulk phase driven by two distinct mechanisms: (1) increasing concentration and (2) decreasing temperature or frustration. The first transition is mediated by a concentration-driven percolation transition of self-limiting, wormlike domains into an intermediate heterogeneous network mesophase, which gradually fills in at high concentration to form a quasiuniform defect bulk state. We find that the percolation threshold is weakly dependent on frustration and shifts to higher concentration as frustration is increased, but depends strongly on the ratio of cohesion to elastic stiffness in the model. The second transition takes place between self-limiting assembly at high-temperature or frustration and phase separation into a condensed bulk state at low temperature or frustration. We consider the competing influences that translational and conformational entropy have on the critical temperature or frustration and show that the self-limiting phase is stabilized at higher frustrations and temperatures than previously expected. Taken together, this understanding of the transition pathways from self-limiting to bulk defect phases of frustrated assembly allows us to map the phase behavior of this 2D minimal model over the full range of concentration. 

This study outlines a probabilistic cyclic shear strain-based procedure for the determination of the minimum shear strain, γcl, required to initiate liquefaction in gravelly soils. The proposed formulation accounts for the influence of void ratio through the shear wave velocity and the grain size distribution through the coefficient of uniformity, Cu. Separate equations for γcl are derived considering four cyclic resistance models that rely on shear wave velocity as a measure of probabilistic liquefaction resistance. Similarities and differences in the resulting γcl for each of these models are identified. The accuracy and uncertainty of cyclic strain-based models in predicting liquefaction in gravelly soils are demonstrated using existing liquefaction case histories where grain size distributions are available. The excess pore pressure response of gravelly soils subjected to earthquake ground motions is evaluated using a subset of the available liquefaction case histories and the cyclic shear strain and energy-based frameworks and is compared to laboratory test specimens. Although the trends in excess pore pressure generation from critical layers in the case histories are comparable to laboratory-based responses, a greater rate of excess pore pressure generation is calculated for the field cases. The models presented in this study can help identify sites that have a high potential for ground failure when used together with other established models.