arXiv subjects

Mitsuhiro Okada

Publications and source records attributed to Mitsuhiro Okada.

12 recordsLinked to original sources

arXiv2026

Evaluation of Deontic Conditional Reasoning in Large Language Models: The Case of Wason's Selection Task

As large language models (LLMs) advance in linguistic competence, their reasoning abilities are gaining increasing attention. In humans, reasoning often performs well in domain specific settings, particularly in normative rather than purely formal contexts. Although prior studies have compared LLM and human reasoning, the domain specificity of LLM reasoning remains underexplored. In this study, we introduce a new Wason Selection Task dataset that explicitly encodes deontic modality to systematically distinguish deontic from descriptive conditionals, and use it to examine LLMs' conditional reasoning under deontic rules. We further analyze whether observed error patterns are better explained by confirmation bias (a tendency to seek rule-supporting evidence) or by matching bias (a tendency to ignore negation and select items that lexically match elements of the rule). Results show that, like humans, LLMs reason better with deontic rules and display matching-bias-like errors. Together, these findings suggest that the performance of LLMs varies systematically across rule types and that their error patterns can parallel well-known human biases in this paradigm.

cs.CL↗

arXiv2026

Abductive Reasoning with Syllogistic Forms in Large Language Models

Research in AI using Large-Language Models (LLMs) is rapidly evolving, and the comparison of their performance with human reasoning has become a key concern. Prior studies have indicated that LLMs and humans share similar biases, such as dismissing logically valid inferences that contradict common beliefs. However, criticizing LLMs for these biases might be unfair, considering our reasoning not only involves formal deduction but also abduction, which draws tentative conclusions from limited information. Abduction can be regarded as the inverse form of syllogism in its basic structure, that is, a process of drawing a minor premise from a major premise and conclusion. This paper explores the accuracy of LLMs in abductive reasoning by converting a syllogistic dataset into one suitable for abduction. It aims to investigate whether the state-of-the-art LLMs exhibit biases in abduction and to identify potential areas for improvement, emphasizing the importance of contextualized reasoning beyond formal deduction. This investigation is vital for advancing the understanding and application of LLMs in complex reasoning tasks, offering insights into bridging the gap between machine and human cognition.

cs.CL↗

arXiv2025

Normative Reasoning in Large Language Models: A Comparative Benchmark from Logical and Modal Perspectives

Normative reasoning is a type of reasoning that involves normative or deontic modality, such as obligation and permission. While large language models (LLMs) have demonstrated remarkable performance across various reasoning tasks, their ability to handle normative reasoning remains underexplored. In this paper, we systematically evaluate LLMs' reasoning capabilities in the normative domain from both logical and modal perspectives. Specifically, to assess how well LLMs reason with normative modals, we make a comparison between their reasoning with normative modals and their reasoning with epistemic modals, which share a common formal structure. To this end, we introduce a new dataset covering a wide range of formal patterns of reasoning in both normative and epistemic domains, while also incorporating non-formal cognitive factors that influence human reasoning. Our results indicate that, although LLMs generally adhere to valid reasoning patterns, they exhibit notable inconsistencies in specific types of normative reasoning and display cognitive biases similar to those observed in psychological studies of human reasoning. These findings highlight challenges in achieving logical consistency in LLMs' normative reasoning and provide insights for enhancing their reliability. All data and code are released publicly at https://github.com/kmineshima/NeuBAROCO.

cs.AI↗

arXiv2024

Exploring Reasoning Biases in Large Language Models Through Syllogism: Insights from the NeuBAROCO Dataset

This paper explores the question of how accurately current large language models can perform logical reasoning in natural language, with an emphasis on whether these models exhibit reasoning biases similar to humans. Specifically, our study focuses on syllogistic reasoning, a form of deductive reasoning extensively studied in cognitive science as a natural form of human reasoning. We present a syllogism dataset called NeuBAROCO, which consists of syllogistic reasoning problems in English and Japanese. This dataset was originally designed for psychological experiments to assess human reasoning capabilities using various forms of syllogisms. Our experiments with leading large language models indicate that these models exhibit reasoning biases similar to humans, along with other error tendencies. Notably, there is significant room for improvement in reasoning problems where the relationship between premises and hypotheses is neither entailment nor contradiction. We also present experimental results and in-depth analysis using a new Chain-of-Thought prompting method, which asks LLMs to translate syllogisms into abstract logical expressions and then explain their reasoning process. Our analysis using this method suggests that the primary limitations of LLMs lie in the reasoning process itself rather than the interpretation of syllogisms.

cs.CL↗

arXiv2023

Evaluating Large Language Models with NeuBAROCO: Syllogistic Reasoning Ability and Human-like Biases

This paper investigates whether current large language models exhibit biases in logical reasoning, similar to humans. Specifically, we focus on syllogistic reasoning, a well-studied form of inference in the cognitive science of human deduction. To facilitate our analysis, we introduce a dataset called NeuBAROCO, originally designed for psychological experiments that assess human logical abilities in syllogistic reasoning. The dataset consists of syllogistic inferences in both English and Japanese. We examine three types of biases observed in human syllogistic reasoning: belief biases, conversion errors, and atmosphere effects. Our findings demonstrate that current large language models struggle more with problems involving these three types of biases.

cs.CL↗

arXiv2021

Sensor-Augmented Egocentric-Video Captioning with Dynamic Modal Attention

Automatically describing video, or video captioning, has been widely studied in the multimedia field. This paper proposes a new task of sensor-augmented egocentric-video captioning, a newly constructed dataset for it called MMAC Captions, and a method for the newly proposed task that effectively utilizes multi-modal data of video and motion sensors, or inertial measurement units (IMUs). While conventional video captioning tasks have difficulty in dealing with detailed descriptions of human activities due to the limited view of a fixed camera, egocentric vision has greater potential to be used for generating the finer-grained descriptions of human activities on the basis of a much closer view. In addition, we utilize wearable-sensor data as auxiliary information to mitigate the inherent problems in egocentric vision: motion blur, self-occlusion, and out-of-camera-range activities. We propose a method for effectively utilizing the sensor data in combination with the video data on the basis of an attention mechanism that dynamically determines the modality that requires more attention, taking the contextual information into account. We compared the proposed sensor-fusion method with strong baselines on the MMAC Captions dataset and found that using sensor data as supplementary information to the egocentric-video data was beneficial, and that our proposed method outperformed the strong baselines, demonstrating the effectiveness of the proposed method.

cs.CV↗

arXiv2020

Dark-state impact on the exciton recombination of WS2 monolayers as revealed by multi-time-scale pump-and-probe spectroscopy

The luminescence yield of transition metal dichalcogenide monolayers frequently suffers from the formation of long-lived dark states, which include excitons with intervalley charge carriers, spin-forbidden transitions, and a large center-of-mass momentum located outside the light cone of dispersion relations. Efficient relaxation from bright exciton states to dark states suppresses the quantum yield of photon emission. In addition, the radiative recombination of excitons is heavily influenced by Auger-type exciton-exciton scattering, which yields another nonradiative relaxation channel at room temperature. Here, we show that Auger-type scattering is promoted not only between (bright) excitons but also between excitons and long-lived dark states. We studied the luminescence dynamics of monolayer WS2 capped with hexagonal BN over broad time ranges of picoseconds to milliseconds using carefully designed pump-and-probe techniques. We observed that luminescence quenching associated with Auger-type scattering occurs on 1-100 microsecond time scales, which thus correspond to the lifetimes of the relevant dark states. The broad distribution of the measured lifetimes implies the impact of various types of long-lived states on the exciton annihilation process.

cond-mat.mes-hall↗

arXiv2020

Microscopic mechanism of van der Waals heteroepitaxy in the formation of MoS2/hBN vertical heterostructures

Recent works have revealed that van der Waals (vdW) epitaxial growth of 2D materials on crystalline substrates, such as hexagonal boron nitride (hBN), leads to formation of self-aligned grains, which results in defect-free stitching between the grains. However, how the weak vdW interaction causes strong limitation on orientation of grains is still not understood yet. In this work, we have focused on investigation of mechanism of self-alignment of MoS2 grains in vdW epitaxial growth on hBN. Through calculation based on density functional theory and the Lennard-Jones potential, we found that interaction energy between MoS2 and hBN strongly depends both on size and orientation of MoS2. We also found that, when size of MoS2 is ca. 40 nm, rotational energy barrier can exceed ~ 1 eV, which should suppress rotation to limit orientation of MoS2 even at growth temperature.

cond-mat.mes-hall↗

arXiv2019

On Quasi Ordinal Diagram Systems

The purposes of this note are the following two; we first generalize Okada-Takeuti's well quasi ordinal diagram theory, utilizing the recent result of Dershowitz-Tzameret's version of tree embedding theorem with gap conditions. Second, we discuss possible use of such strong ordinal notation systems for the purpose of a typical traditional termination proof method for term rewriting systems, especially for second-order (pattern-matching-based) rewriting systems including a rewrite-theoretic version of Buchholz's hydra game.

cs.LO↗

arXiv2018

A Note on Switching Conditions for the Generalized Logical Connectives in Multiplicative Linear Logic

Danos and Regnier (1989) introduced the par-switching condition for multiplicative proof-structures and simplified the sequentialization theorem of Girard (1987) by means of par-switching. Danos and Regnier (1989) also generalized the par-switching to a switching for n-ary connectives (hereafter called an n-ary switching) and showed that the "expansion" property holds, namely that any "excluded-middle" formula admits a correct proof-net in the sense of their n-ary switching. They added a remark that the sequentialization theorem does not hold with their switching. Their definition of switching for n-ary connectives is a natural generalization of the original switching for the binary connectives. However, there are many other possible definitions of switching for n-ary connectives. We give an alternative and "natural" definition of n-ary switching, and we show that the proof of sequentialization theorem by Olivier Laurent with the par-switching works for our n-ary switching; Consequently, the sequentialization theorem holds for our n-ary switching. On the other hand, we remark that the "expansion" property no longer holds under our switching anymore. We point out that no definition of n-ary switching satisfies both the sequentialization theorem and the "expansion" property at the same time except for the purely tensor-based (or purely par-based) connectives.

math.LO↗

arXiv2013

Inductive-data-type Systems

In a previous work ("Abstract Data Type Systems", TCS 173(2), 1997), the last two authors presented a combined language made of a (strongly normalizing) algebraic rewrite system and a typed lambda-calculus enriched by pattern-matching definitions following a certain format, called the "General Schema", which generalizes the usual recursor definitions for natural numbers and similar "basic inductive types". This combined language was shown to be strongly normalizing. The purpose of this paper is to reformulate and extend the General Schema in order to make it easily extensible, to capture a more general class of inductive types, called "strictly positive", and to ease the strong normalization proof of the resulting system. This result provides a computation model for the combination of an algebraic specification language based on abstract data types and of a strongly typed functional language with strictly positive inductive types.

cs.LO↗

arXiv2008

The Calculus of Algebraic Constructions

This paper is concerned with the foundations of the Calculus of Algebraic Constructions (CAC), an extension of the Calculus of Constructions by inductive data types. CAC generalizes inductive types equipped with higher-order primitive recursion, by providing definitions of functions by pattern-matching which capture recursor definitions for arbitrary non-dependent and non-polymorphic inductive types satisfying a strictly positivity condition. CAC also generalizes the first-order framework of abstract data types by providing dependent types and higher-order rewrite rules.

cs.LO↗