25 papers · ranked by Valyu relevance
Priyanka Kargupta, Shuyue Stella Li, Haocheng Wang, Jinu Lee + 8 more
Large language models (LLMs) solve complex problems yet fail on simpler variants, suggesting they achieve correct outputs through mechanisms fundamentally different from human reasoning. To understand this gap, we synthesize cognitive science research into a taxonomy of 28 cognitive elements spanning reasoning…
Qiguang Chen, Jinhao Liu, Qin Li, Yimeng Zhang + 17 more
Understanding how information is dynamically accumulated and transformed in human reasoning has long challenged cognitive psychology, philosophy, and artificial intelligence. Existing accounts, from classical logic to probabilistic models, illuminate aspects of output or individual modelling, but do not offer a…
Reto Gubelmann
Taking Leibniz' ideal of a universal truth-calculating machine as a vantage point, this article provides a philosophically sound analysis of the concept of reasoning in NLP. It argues that reasoning always involves inference, which in turn requires being guided by reason relations. Based on this, the article argues…
Sebastijan Veselic, Nour Mohsen, Lennart Luettgau, Elena Gutierrez + 5 more
Reasoning flexibly composes known elements to solve novel problems. Recent theories suggest the brain uses the axis of time to compose elements for reasoning. In this view, elements are packaged into fast neural sequences, with each sequence exploring the implications of a different composition. Using…
Hendrik Kempt, Alon Lavie
Reasoning has long been understood as a pathway between stages of understanding. Proper reasoning leads to understanding of a given subject. This reasoning was conceptualized as a process of understanding in a particular way, i.e., "symbolic reasoning". Foundational Models (FM) demonstrate that this is not a necessary…
Robert Ricco, Stefan Bogaerts
Argumentative reasoning (AR) refers to the kind of reasoning used when individuals engage in argument about a disputed claim or proposed action. In its mature, most proficient form, AR involves several reasoning skills such as providing effective justification for one’s claims, anticipating and defending against…
Tao Hong, William R. Stauffer
Complex economic decisions are often combinatorial: they require individuals to select from many alternatives under strict constraints on time, resources, and energy. Combinatorial reasoning is the cognitive process that enables decision makers to construct and evaluate multiple potential solutions in the face of these…
Luke J. Hearne, Conor Robinson, Luca Cocchi, Takuya Ito
Relational reasoning—the capacity to understand how elements relate to one another—is a defining feature of human intelligence, yet its computational basis remains unclear. Here, we combined human neuroimaging (7T fMRI) and artificial neural network modeling to examine relational reasoning in biological and artificial…
Mothilal, Ramaravind Kommiya, Zhang, Sally + 4 more
HCI literature on LLMs often follows a tokenistic mention of LLM reasoning for various reasons, including for critiquing LLMs' unreliability or for studying users' perceptions regarding a specific task. We find that the reasoning capabilities and behaviors of LLMs are explicitly mentioned in 63% (N=162) of the papers…
Geoffrey Whittle-Walls
Human reasoning is traditionally modeled through rational-order frameworks that assume stability, separability, and coherence. Yet across judgment, valuation, perception, and social decision-making, empirical work consistently reveals patterned violations of these assumptions. These deviations intensify in real-world…
Xuanqiao Lin, Yizhou Yang, Yuecheng Ren
Recent advances in deep-reasoning large language models (LLMs)-including OpenAI's GPT series and open-source DeepSeek models-have expanded their potential applications in ophthalmology. In ophthalmology, image interpretation continues to rely primarily on conventional computer vision and vision language model…
Anna A. Ivanova, Carina Kauf, Ruimin Gao, Jingyuan Selena She + 6 more
The brain’s language network is often implicated in the representation and manipulation of abstract semantic knowledge. However, this view is inconsistent with a large body of evidence suggesting that language processing is neurally distinct from the rest of cognition. Here, we use precision brain imaging to uncover a…
Zixuan Wang, Yuanyuan Lei
Logical reasoning serve as a central capability in LLMs and includes three main forms: deductive, inductive, and abductive reasoning. In this work, we study the knowledge representations of these reasoning types in LLMs and analyze the correlations among them. Our analysis shows that each form of logical reasoning can…
Isaac R Christian, Samuel A Nastase
Humans spend considerable time contemplating the minds of others. But this ability is not limited to external agents-we also turn the lens for reading minds inward, reflecting on our own thoughts, emotions, and sense of self. Some processes involved in reasoning about minds may rely on shared mechanisms, while others…
Kyle Fiore Law, Stylianos Syropoulos, Paige Amormino, Abigail Marsh + 3 more
Humans can care about distant strangers, an adaptive advantage that enables our species to cooperate in increasingly large-scale groups. Theoretical frameworks accounting for an expansive moral circle and altruistic behavior are often framed as a dichotomy between competing pathways of emotion-driven empathy versus…
Zhimin Hu, Beatriz Martín-Luengo, Eduardo Navarrete
This research investigates the moral Foreign Language Effect (mFLE) from a metacognitive perspective. Grounded in the Dual-Process framework, previous research posits using a foreign language evokes more utilitarianism by dampening emotional responses and promoting analytical reasoning. However, the role of…
Paulo Pirozelli, Victor Hugo Nascimento Rocha, Fabio G. Cozman, Douglas Aldred
Arguments are a fundamental aspect of human reasoning, in which claims are supported, challenged, and weighed against one another. We present an end-to-end large language model (LLM)-based system for reconstructing arguments from natural language text into abstract argument graphs. The system follows a multi-stage…
Authors not listed
Incorporating prior domain knowledge into Bayesian optimization (BO) remains difficult for statistical methods, which also typically suffer from limited interpretability. Large language models (LLMs) offer complementary strengths in reasoning and knowledge integration, but it remains unclear when and how they improve…
Authors not listed
Generative artificial intelligence (AI) tools such as large language models (LLMs) have become ubiquitous in everyday life, and are also increasingly finding applications in the chemical sciences. Although LLMs have achieved impressive performance on many chemistry tasks, optimal performance requires proper use…
Jacob A. Parker, Alexandre L.S. Filipowicz, Kristen Li, Vijay Balasubramanian + 2 more
Human decision-making behavior varies widely across individuals and task conditions. This variability is often interpreted in terms of different suboptimal decision strategies, but the principles that govern these suboptimalities remain poorly understood. We propose that some of these suboptimalities can be understood…
Authors not listed
Realizing the promise of artificial intelligence (AI) to accelerate scientific progress and deliver technological impact depends on how effectively AI can be integrated into real-world decision- making processes. As Peter Norvig states, “Somewhat remarkably, almost all AI research until very recently has assumed that…
Authors not listed
This paper introduces Post-Inspection Analysis (PIA), a concept for understanding and reverse engineering balanced equations of similar reactions. Here we explore the idea that balanced equations contain useable information observed by grouping and analysing similar equations, allowing us to uncover observations that…
Authors not listed
Machine learning is increasingly used to predict reaction properties such as barrier heights, reaction energies, rates, or yields, as well as the underlying molecular geometries, including transition state structures. While such predictions have the potential to provide mechanistic insight for high-impact applications…
Mengmin Xu, Yan Ren
Building upon foundational psychological theories of event segmentation, this study addresses the limitation of overreliance on temporal boundaries as the primary segmentation criterion. Drawing on two experiments of direct and indirect causation in Mandarin Chinese, this study demonstrates how cognitive segmentation…
Authors not listed
The value of generative artificial intelligence (AI) for teaching and learning is currently hotly debated. Concerns regarding the accuracy of information produced by generative AI as well as student over-reliance on this tool coexist with excitement about tailored opportunities that AI may provide for educational…