20 papers · ranked by Valyu relevance
Davide Maltoni, Ferrara, Matteo
Recent large language models have demonstrated relevant capabilities in solving problems that require logical reasoning; however, the corresponding internal mechanisms remain largely unexplored. In this paper, we show that a small language model can solve a deductive reasoning task by learning the underlying rules…
信之 和泉, Jiafeng Xiong, Marco Valentino, André Freitas
Attributional inference, the ability to predict latent intentions behind observed actions, is a critical yet underexplored capability for large language models (LLMs) operating in multiagent environments. Traditional natural language inference (NLI), in fact, fails to capture the nuanced, intention-driven reasoning…
Robbe Van den Eede, Marc Denecker
Inductive definitions are an important form of knowledge. The logic FO(ID) is an extension of classical first-order logic FO with general non-monotone inductive definitions. Most existing proof systems for inductive definitions impose syntactic constraints on their definitions, thereby excluding many useful and natural…
Pierre Collet, Felipe Quezada-Diaz, Carla Taramasco, Shing Wai Wong + 2 more
Title: Simple Summary Traditional single-test strategies using faecal immunochemical tests (FIT) for the detection of colorectal cancer can often yield uncertain results, leading to unnecessary colonoscopies or missed diagnoses. In this paper, we show how when applied to four consecutive FITs, Updated Bayesian…
Judd T. Kalkhoven
Athletic injury remains inadequately conceptualised and poorly defined. Existing definitions lack the conceptual clarity and logical coherence required to support its maturation into a scientifically meaningful and reliably investigable concept. Related constructs that are often integrated into various operational…
Jianfeng Xu
We study lossy compression of a finite statement source generated in a fixed deductive environment. The source symbols are statements in a knowledge base endowed with a proof system, and reconstruction fidelity is measured by preservation of deductive closure rather than by symbolwise equality. This induces, once the…
Matthew Collins, Timo Eckhardt, David Pym
Information is one of the most widely-discussed concepts of the current era. However, a great deal of insightful work notwithstanding, it is yet to be given wholly convincing logical or mathematical foundations. Without them, we lack adequate reasoning tools for understanding the complex ecosystems of systems upon…
Alfredo Verde, Emanuela Saveria Gritti
Forensic psychological and psychiatric evaluations-understood here as assessments addressing legally operative questions such as, for instance, competency to stand trial, criminal responsibility (sanity), violence risk, but also those on legal capacity and child custody-must meet stringent standards of scientific…
Tommaso Costa
In this paper, I argue that the problem of induction dissolves when recast in terms of logical coherence (understood as internal consistency of credences under updating) rather than truth. Following E. T. Jaynes, probability is interpreted not as frequency or decision rule but as the extension of deductive logic to…
Deniz Özyıldız, Ciyang Qing, Floris Roelofsen, Wataru Uegaki + 1 more
This paper evaluates empirical methods for detecting the focus-sensitivity of clause-embedding predicates, with the objective of making them applicable to predicates within and across languages. Our positive proposal is a test based on a combination of inferential judgments and truth-value judgments. We compare this…
Ivano Ciardelli
We introduce inquisitive action logic, InqAL, a multi-agent modal logic for reasoning about action. While traditional approaches focus on what properties of the outcome an agent can force, InqAL also captures what aspects of the outcome an agent determines through their actions. As we argue, such claims of agentive…
Chuanyong Xu, Ning Mei, Wenshan Dong, Rongcheng Hu + 2 more
Flexible adaptation requires inferring the causes of feedback and adjusting accordingly. When tasks are hierarchically structured with high-level rules governing low-level perceptual judgements, good performance requires hierarchical reasoning. Yet the computational and neural mechanisms underlying this process remain…
Robert Hoehndorf, Paul N. Schofield, Georgios V. Gkoutos
Computational physiology needs representations that connect traits across biological scales while distinguishing causal, constitutive, and mathematical dependencies. We present PhysioMap, an ontology-grounded knowledge base of contextualized physiological traits and precisely defined relation types. A versioned…
Jiayu Han, Yanfei Zhang, Jon Andoni Duñabeitia
This study conducts a case study of liǎnhóng (脸红) to examine how inference works in the diachronic change in conversational implicature. It proposes three types of inference, namely ad hoc inference, entrenched inference, and conventionalized inference, with each contributing distinctively to semantic change. The…
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…
Martin Smith
The paper is concerned with a special class of inferences, in which we draw conclusions about individual people based on evidence about the groups to which they belong. One thing that is notable about these inferences is that they are often subject to a kind of moral criticism. By judging people in this way, it is…
Ulisses Franceschi Eliano
In his 1996 doctoral thesis, Maurice Pagnucco created the first AGM-like abductive expansion operation. Taking his operation as a basis, as well as a taxonomy -- inspired by Atocha Aliseda -- responsible for highlighting and formalizing the main components of abductive reasoning, the main aim of this paper is to…
Shiyue Li, Yiling Wang, Zhanpeng Shu, Ramon Grima + 2 more
Biochemical reactions are inherently stochastic, with their kinetics commonly described by chemical master equations (CMEs). However, the discrete nature of molecular states renders likelihood-based parameter inference from CMEs computationally intensive. Here, we introduce an inference method that leverages analytical…
Emma Clephas, Bente Raaso, Sara Spolaore, Lars Schwabe + 1 more
The integration of separate memories sharing overlapping elements allows for inferring relationships between events that have never been experienced together. Recent findings indicate that stress impairs this process of mnemonic inference, which may have important implications for stress-related psychopathologies.…
Authors not listed
Artificial intelligence (AI) is reshaping chemical engineering. Still, its role in safety-critical operations is limited because we rarely see tools that link physical models with data-driven methods. This study brings together three elements: physics-constrained neural networks, uncertainty quantification, and a…