arXiv subjects

Masahiro Suzuki

Publications and source records attributed to Masahiro Suzuki.

At least 37 records · Page 2Linked to original sources

arXiv2025

Agent-Based Simulation of a Financial Market with Large Language Models

In real-world stock markets, certain chart patterns -- such as price declines near historical highs -- cannot be fully explained by fundamentals alone. These phenomena suggest the presence of path dependence in price formation, where investor decisions are influenced not only by current market conditions but also by the trajectory of prices leading up to the present. Path dependence has drawn attention in behavioral finance as a key mechanism behind such anomalies. One plausible driver of path dependence is human loss aversion, anchored to individual reference points like purchase prices or past peaks, which vary with personal context. However, capturing such subtle behavioral tendencies in traditional agent-based market simulations has remained a challenge. We propose the Fundamental-Chartist-LLM-Agent (FCLAgent), which uses large language models (LLMs) to emulate human-like trading decisions. In this framework, (1) buy/sell decisions are made by LLMs based on individual situations, while (2) order price and volume follow standard rule-based methods. Simulations show that FCLAgents reproduce path-dependent patterns that conventional agents fail to capture. Furthermore, an analysis of FCLAgents' behavior reveals that the reference points guiding loss aversion vary with market trajectories, highlighting the potential of LLM-based agents to model nuanced investor behavior.

cs.CE↗

arXiv2025

Fe contribution to the magnetic anisotropy of $L{1_0}$-ordered FePt thin films studied by angle-dependent x-ray magnetic circular dichroism

Among magnetic thin films with perpendicular magnetic anisotropy (PMA), $L1_0$-ordered FePt has attracted significant attention because of its exceptionally strong PMA. However, the microscopic origin of its strong PMA has not been elucidated experimentally. We have investigated the contribution of the Fe $3d$ electrons to its magnetic anisotropy energy by angle-dependent x-ray magnetic circular dichroism at the Fe $L_{2,3}$ edge. By this technique, one can deduce the magnetic dipole moment $m_\text{T}$, which represents the anisotropic spatial distribution of spin-polarized electrons, and the orbital moment anisotropy (OMA) of Fe $3d$ electrons. Detected finite $m_\text{T}$ indicates that the spin-polarized Fe $3d$ electrons are distributed preferentially in the out-of-plane direction of the films. This $m_\text{T}$ of Fe overwhelms the positive contribution of OMA to PMA, and reduces the PMA of $L1_0$-ordered FePt thin films, consistent with a previous first-principles calculation. The present result implies that a large positive contribution of the non-magnetic element Pt rather than Fe governs the PMA of $L1_0$-ordered FePt thin films.

cond-mat.mtrl-sci↗

arXiv2025

Inference-Time Text-to-Video Alignment with Diffusion Latent Beam Search

The remarkable progress in text-to-video diffusion models enables the generation of photorealistic videos, although the content of these generated videos often includes unnatural movement or deformation, reverse playback, and motionless scenes. Recently, an alignment problem has attracted huge attention, where we steer the output of diffusion models based on some measure of the content's goodness. Because there is a large room for improvement of perceptual quality along the frame direction, we should address which metrics we should optimize and how we can optimize them in the video generation. In this paper, we propose diffusion latent beam search with lookahead estimator, which can select a better diffusion latent to maximize a given alignment reward at inference time. We then point out that improving perceptual video quality with respect to alignment to prompts requires reward calibration by weighting existing metrics. This is because when humans or vision language models evaluate outputs, many previous metrics to quantify the naturalness of video do not always correlate with the evaluation. We demonstrate that our method improves the perceptual quality evaluated on the calibrated reward, VLMs, and human assessment, without model parameter update, and outputs the best generation compared to greedy search and best-of-N sampling under much more efficient computational cost. The experiments highlight that our method is beneficial to many capable generative models, and provide a practical guideline: we should prioritize the inference-time compute allocation into enabling the lookahead estimator and increasing the search budget, rather than expanding the denoising steps.

cs.CV↗

arXiv2025

Difficulty-Aware Score Generation for Piano Sight-Reading

Adapting learning materials to the level of skill of a student is important in education. In the context of music training, one essential ability is sight-reading -- playing unfamiliar scores at first sight -- which benefits from progressive and level-appropriate practice. However, creating exercises at the appropriate level of difficulty demands significant time and effort. We address this challenge as a controlled symbolic music generation task that aims to produce piano scores with a desired difficulty level. Controlling symbolic generation through conditioning is commonly done using control tokens, but these do not always have a clear impact on global properties, such as difficulty. To improve conditioning, we introduce an auxiliary optimization target for difficulty prediction that helps prevent conditioning collapse -- a common issue in which models ignore control signals in the absence of explicit supervision. This auxiliary objective helps the model to learn internal representations aligned with the target difficulty, enabling more precise and adaptive score generation. Evaluation with automatic metrics and expert judgments shows better control of difficulty and potential educational value. Our approach represents a step toward personalized music education through the generation of difficulty-aware practice material.

cs.SD↗

arXiv2025

Generative Emergent Communication: Large Language Model is a Collective World Model

Large Language Models (LLMs) have demonstrated a remarkable ability to capture extensive world knowledge, yet how this is achieved without direct sensorimotor experience remains a fundamental puzzle. This study proposes a novel theoretical solution by introducing the Collective World Model hypothesis. We argue that an LLM does not learn a world model from scratch; instead, it learns a statistical approximation of a collective world model that is already implicitly encoded in human language through a society-wide process of embodied, interactive sense-making. To formalize this process, we introduce generative emergent communication (Generative EmCom), a framework built on the Collective Predictive Coding (CPC). This framework models the emergence of language as a process of decentralized Bayesian inference over the internal states of multiple agents. We argue that this process effectively creates an encoder-decoder structure at a societal scale: human society collectively encodes its grounded, internal representations into language, and an LLM subsequently decodes these symbols to reconstruct a latent space that mirrors the structure of the original collective representations. This perspective provides a principled, mathematical explanation for how LLMs acquire their capabilities. The main contributions of this paper are: 1) the formalization of the Generative EmCom framework, clarifying its connection to world models and multi-agent reinforcement learning, and 2) its application to interpret LLMs, explaining phenomena such as distributional semantics as a natural consequence of representation reconstruction. This work provides a unified theory that bridges individual cognitive development, collective language evolution, and the foundations of large-scale AI.

cs.AI↗

arXiv2025

System 0/1/2/3: Quad-process theory for multi-timescale embodied collective cognitive systems

This paper introduces the System 0/1/2/3 framework as an extension of dual-process theory, employing a quad-process model of cognition. Expanding upon System 1 (fast, intuitive thinking) and System 2 (slow, deliberative thinking), we incorporate System 0, which represents pre-cognitive embodied processes, and System 3, which encompasses collective intelligence and symbol emergence. We contextualize this model within Bergson's philosophy by adopting multi-scale time theory to unify the diverse temporal dynamics of cognition. System 0 emphasizes morphological computation and passive dynamics, illustrating how physical embodiment enables adaptive behavior without explicit neural processing. Systems 1 and 2 are explained from a constructive perspective, incorporating neurodynamical and AI viewpoints. In System 3, we introduce collective predictive coding to explain how societal-level adaptation and symbol emergence operate over extended timescales. This comprehensive framework ranges from rapid embodied reactions to slow-evolving collective intelligence, offering a unified perspective on cognition across multiple timescales, levels of abstraction, and forms of human intelligence. The System 0/1/2/3 model provides a novel theoretical foundation for understanding the interplay between adaptive and cognitive processes, thereby opening new avenues for research in cognitive science, AI, robotics, and collective intelligence.

cs.AI↗

arXiv2025

Ionized Gas in an Annular Region

We consider a plasma that is created by a high voltage difference $λ$, which is known as a Townsend discharge. We consider it to be confined to the region $Ω$ between two concentric spheres, two concentric cylinders, or more generally between two star-shaped surfaces. We first prove that if the plasma is initially relatively dilute, then either it may remain dilute for all time or it may not, depending on a certain parameter $κ(λ, Ω)$. Secondly, we prove that there is a connected one-parameter family of steady states. This family connects the non-ionized gas to a plasma, either with a sparking voltage $λ^*$ or with very high ionization, at least in the cylindrical or spherical cases.

math.AP↗

arXiv2025

Economy Watchers Survey Provides Datasets and Tasks for Japanese Financial Domain

Natural language processing (NLP) tasks in English and general domains are widely available and are often used to evaluate pre-trained language models. In contrast, fewer tasks are available for languages other than English and in the financial domain. Particularly, tasks in the Japanese and financial domains are limited. We develop two large datasets using data published by a Japanese central government agency. The datasets provide three Japanese financial NLP tasks, including 3- and 12-class classifications for categorizing sentences, along with a 5-class classification task for sentiment analysis. Our datasets are designed to be comprehensive and updated by leveraging an automatic update framework that ensures that the latest task datasets are publicly always available.

cs.CL↗

arXiv2024

Refined and Segmented Price Sentiment Indices from Survey Comments

We aim to enhance a price sentiment index and to more precisely understand price trends from the perspective of not only consumers but also businesses. We extract comments related to prices from the Economy Watchers Survey conducted by the Cabinet Office of Japan and classify price trends using a large language model (LLM). We classify whether the survey sample reflects the perspective of consumers or businesses, and whether the comments pertain to goods or services by utilizing information on the fields of comments and the industries of respondents included in the Economy Watchers Survey. From these classified price-related comments, we construct price sentiment indices not only for a general purpose but also for more specific objectives by combining perspectives on consumers and prices, as well as goods and services. It becomes possible to achieve a more accurate classification of price directions by employing a LLM for classification. Furthermore, integrating the outputs of multiple LLMs suggests the potential for the better performance of the classification. The use of more accurately classified comments allows for the construction of an index with a higher correlation to existing indices than previous studies. We demonstrate that the correlation of the price index for consumers, which has a larger sample size, is further enhanced by selecting comments for aggregation based on the industry of the survey respondents.

cs.CL↗

arXiv2024

ADOPT: Modified Adam Can Converge with Any $β_2$ with the Optimal Rate

Adam is one of the most popular optimization algorithms in deep learning. However, it is known that Adam does not converge in theory unless choosing a hyperparameter, i.e., $β_2$, in a problem-dependent manner. There have been many attempts to fix the non-convergence (e.g., AMSGrad), but they require an impractical assumption that the gradient noise is uniformly bounded. In this paper, we propose a new adaptive gradient method named ADOPT, which achieves the optimal convergence rate of $\mathcal{O} ( 1 / \sqrt{T} )$ with any choice of $β_2$ without depending on the bounded noise assumption. ADOPT addresses the non-convergence issue of Adam by removing the current gradient from the second moment estimate and changing the order of the momentum update and the normalization by the second moment estimate. We also conduct intensive numerical experiments, and verify that our ADOPT achieves superior results compared to Adam and its variants across a wide range of tasks, including image classification, generative modeling, natural language processing, and deep reinforcement learning. The implementation is available at https://github.com/iShohei220/adopt.

cs.LG↗

arXiv2024

Object-Centric Temporal Consistency via Conditional Autoregressive Inductive Biases

Unsupervised object-centric learning from videos is a promising approach towards learning compositional representations that can be applied to various downstream tasks, such as prediction and reasoning. Recently, it was shown that pretrained Vision Transformers (ViTs) can be useful to learn object-centric representations on real-world video datasets. However, while these approaches succeed at extracting objects from the scenes, the slot-based representations fail to maintain temporal consistency across consecutive frames in a video, i.e. the mapping of objects to slots changes across the video. To address this, we introduce Conditional Autoregressive Slot Attention (CA-SA), a framework that enhances the temporal consistency of extracted object-centric representations in video-centric vision tasks. Leveraging an autoregressive prior network to condition representations on previous timesteps and a novel consistency loss function, CA-SA predicts future slot representations and imposes consistency across frames. We present qualitative and quantitative results showing that our proposed method outperforms the considered baselines on downstream tasks, such as video prediction and visual question-answering tasks.

cs.CV↗

arXiv2024

Stationary flows for viscous heat-conductive fluid in a perturbed half-space

We consider the non-isentropic compressible Navier-Stokes equation in a perturbed half space with an outflow boundary condition as well as the supersonic condition. This equation models a compressible viscous, heat-conductive, and Newtonian polytropic fluid. We show the unique existence of stationary solutions for the perturbed half-space. The stationary solution constructed in this paper depends on all directions and has multidirectional flow. We also prove the asymptotic stability of this stationary solution.

math.AP↗

arXiv2024

Enhancing Unimodal Latent Representations in Multimodal VAEs through Iterative Amortized Inference

Multimodal variational autoencoders (VAEs) aim to capture shared latent representations by integrating information from different data modalities. A significant challenge is accurately inferring representations from any subset of modalities without training an impractical number (2^M) of inference networks for all possible modality combinations. Mixture-based models simplify this by requiring only as many inference models as there are modalities, aggregating unimodal inferences. However, they suffer from information loss when modalities are missing. Alignment-based VAEs address this by aligning unimodal inference models with a multimodal model through minimizing the Kullback-Leibler (KL) divergence but face issues due to amortization gaps, which compromise inference accuracy. To tackle these problems, we introduce multimodal iterative amortized inference, an iterative refinement mechanism within the multimodal VAE framework. This method overcomes information loss from missing modalities and minimizes the amortization gap by iteratively refining the multimodal inference using all available modalities. By aligning unimodal inference to this refined multimodal posterior, we achieve unimodal inferences that effectively incorporate multimodal information while requiring only unimodal inputs during inference. Experiments on benchmark datasets show that our approach improves inference performance, evidenced by higher linear classification accuracy and competitive cosine similarity, and enhances cross-modal generation, indicated by lower FID scores. This demonstrates that our method enhances inferred representations from unimodal inputs.

cs.LG↗

arXiv2024

On the relationship between orbital moment anisotropy, magnetocrystalline anisotropy, and Dzyaloshinskii-Moriya interaction in W/Co/Pt trilayers

We have studied the Co layer thickness dependences of magnetocrystalline anisotropy (MCA), Dzyaloshinskii-Moriya interaction (DMI), and orbital moment anisotropy (OMA) in W/Co/Pt trilayers, in order to clarify their correlations with each other. We find that the MCA favors magnetization along the film normal and monotonically increases with decreasing effective magnetic layer thickness ($t_\mathrm{eff}$). The magnitude of the Dzyaloshinskii-Moriya exchange constant ($|D|$) increases with decreasing $t_\mathrm{eff}$ until $t_\mathrm{eff} \sim$1 nm, below which $|D|$ decreases. The MCA and $|D|$ scale with $1/t_\mathrm{eff}$ for $t_\mathrm{eff}$ larger than $\sim$1 nm, indicating an interfacial origin. The increase of MCA with decreasing $t_\mathrm{eff}$ continues below $t_\mathrm{eff}$ $\sim$ 1 nm, but with a slower rate. To clarify the cause of the $t_\mathrm{eff}$ dependences of MCA and DMI, the OMA of Co in W/Co/Pt trilayers is studied using x-ray magnetic circular dichroism (XMCD). We find non-zero OMA when $t_\mathrm{eff}$ is smaller than $\sim$0.8 nm. The OMA increases with decreasing $t_\mathrm{eff}$ more rapidly than what is expected from the MCA, indicating that factors other than OMA contribute to the MCA at small $t_\mathrm{eff}$. The $t_\mathrm{eff}$ dependence of the OMA also suggests that $|D|$ at $t_\mathrm{eff}$ smaller than $\sim$1 nm is not related to the OMA at the interface. We propose that the growth of Co on W results in a strain and/or texture that reduces the interfacial DMI, and, to some extent, MCA at small $t_\mathrm{eff}$.

cond-mat.mtrl-sci↗

arXiv2024

Interactive DualChecker for Mitigating Hallucinations in Distilling Large Language Models

Large Language Models (LLMs) have demonstrated exceptional capabilities across various machine learning (ML) tasks. Given the high costs of creating annotated datasets for supervised learning, LLMs offer a valuable alternative by enabling effective few-shot in-context learning. However, these models can produce hallucinations, particularly in domains with incomplete knowledge. Additionally, current methods for knowledge distillation using LLMs often struggle to enhance the effectiveness of both teacher and student models. To address these challenges, we introduce DualChecker, an innovative framework designed to mitigate hallucinations and improve the performance of both teacher and student models during knowledge distillation. DualChecker employs ContextAligner to ensure that the context provided by teacher models aligns with human labeling standards. It also features a dynamic checker system that enhances model interaction: one component re-prompts teacher models with more detailed content when they show low confidence, and another identifies borderline cases from student models to refine the teaching templates. This interactive process promotes continuous improvement and effective knowledge transfer between the models. We evaluate DualChecker using a green innovation textual dataset that includes binary, multiclass, and token classification tasks. The experimental results show that DualChecker significantly outperforms existing state-of-the-art methods, achieving up to a 17% improvement in F1 score for teacher models and 10% for student models. Notably, student models fine-tuned with LLM predictions perform comparably to those fine-tuned with actual data, even in a challenging domain. We make all datasets, models, and code from this research publicly available.

cs.CL↗

arXiv2024

JaFIn: Japanese Financial Instruction Dataset

We construct an instruction dataset for the large language model (LLM) in the Japanese finance domain. Domain adaptation of language models, including LLMs, is receiving more attention as language models become more popular. This study demonstrates the effectiveness of domain adaptation through instruction tuning. To achieve this, we propose an instruction tuning data in Japanese called JaFIn, the Japanese Financial Instruction Dataset. JaFIn is manually constructed based on multiple data sources, including Japanese government websites, which provide extensive financial knowledge. We then utilize JaFIn to apply instruction tuning for several LLMs, demonstrating that our models specialized in finance have better domain adaptability than the original models. The financial-specialized LLMs created were evaluated using a quantitative Japanese financial benchmark and qualitative response comparisons, showing improved performance over the originals.

cs.CL↗

arXiv2024

Commissioning of a compact multibend achromat lattice: A new 3 GeV synchrotron radiation facility

NanoTerasu, a new 3 GeV synchrotron light source in Japan, began user operation in April 2024. It provides high-brilliance soft to tender X-rays and covers a wide spectral range from ultraviolet to tender X-rays. Its compact storage ring with a circumference of 349 m is based on a four-bend achromat lattice to provide two straight sections in each cell for insertion devices with a natural horizontal emittance of 1.14 nm rad, which is small enough for soft X-rays users. The NanoTerasu accelerator incorporates several innovative technologies, including a full-energy injector C-band linear accelerator with a length of 110 m, an in-vacuum off-axis injection system, a four-bend achromat with B-Q combined bending magnets, and a TM020 mode accelerating cavity with built-in higher-order-mode dampers in the storage ring. This paper presents the accelerator machine commissioning over a half-year period and our model-consistent ring optics correction. The first user operation with a stored beam current of 160 mA is also reported. We summarize the storage ring parameters obtained from the commissioning. This is helpful for estimating the effective optical properties of synchrotron radiation at NanoTerasu.

physics.acc-ph↗

arXiv2024

The Embodied World Model Based on LLM with Visual Information and Prediction-Oriented Prompts

In recent years, as machine learning, particularly for vision and language understanding, has been improved, research in embedded AI has also evolved. VOYAGER is a well-known LLM-based embodied AI that enables autonomous exploration in the Minecraft world, but it has issues such as underutilization of visual data and insufficient functionality as a world model. In this research, the possibility of utilizing visual data and the function of LLM as a world model were investigated with the aim of improving the performance of embodied AI. The experimental results revealed that LLM can extract necessary information from visual data, and the utilization of the information improves its performance as a world model. It was also suggested that devised prompts could bring out the LLM's function as a world model.

cs.AI↗