arXiv subjects

Takehisa Yairi

Publications and source records attributed to Takehisa Yairi.

16 recordsLinked to original sources

arXiv2026

Tsubame: Tree Replay for Diffusion-Based Speculative Decoding

Context-aware dynamic trees allocate the speculative decoding budget according to draft path probabilities, adapting their depth and branching to the current context. Under stochastic decoding, however, we find that this structural advantage does not always compensate for the acceptance gains of random sampling paired with advanced verification, and such dynamic trees can fall behind sampled chains in some settings. These trees grow their topology from the candidates themselves, so the tokens submitted for verification are typically the deterministic high-score tokens selected during construction. This coupling is not inherent: once the topology is fixed, its nodes can be repopulated by sampling, allowing dynamic trees to retain their structural advantage while also benefiting from random sampling and advanced verification. Diffusion-based drafters make this practical, as their parallel outputs or lightweight conditional corrections allow candidates to be regenerated cheaply after the complete topology is known. We introduce Tsubame, a two-pass tree speculative decoding framework for diffusion-based drafters. The first pass plans and freezes a context-aware topology using draft path scores; the second replays the fixed topology, sampling the tokens that populate its nodes to form the candidate tree for verification. We prove that Tsubame is lossless under compatible sampling and verification strategies. Experiments across three diffusion-based drafters, six datasets, and multiple candidate budgets show that Tsubame improves acceptance length and throughput over deterministic trees, including settings where it reverses their disadvantage against sampled chains.

cs.CL↗

arXiv2026

RheoSampling: Resolving the One-Hot Dilemma in Stochastic Dynamic-Tree Speculative Decoding

Speculative decoding accelerates LLM inference by drafting multiple tokens in parallel, with tree-based methods further improving efficiency through hierarchical structures. Dynamic-tree methods such as EAGLE-3 perform well under greedy decoding via deterministic top-K expansion and global pruning. However, in stochastic decoding (T>0), this mechanism collapses the draft distribution into one-hot probabilities, causing a severe drop in acceptance rate. This creates a dilemma: dynamic-tree methods sacrifice stochastic sampling to preserve context-aware topology, while static-tree methods preserve stochastic sampling with context-agnostic structures. The issue arises because the same probability distribution is used for two conflicting tasks: constructing the tree and verifying tokens. This coupling makes direct injection of randomness challenging due to the resulting stochastic process. We resolve this by decoupling these roles: RheoSampling assigns a token sampled from the draft distribution a proxy probability for tree expansion and pruning alongside its true sampling probability for verification. Specifically, we inject a sampled token among the deterministic top-K slots and treat it with different probabilities during construction and verification, making RheoSampling the first dynamic-tree method with both context-aware top-K construction and stochastic sampling while maintaining losslessness. We establish the lossless guarantee through an equivalence-class analysis that compresses the stochastic tree space into tractable classes. An OT-based verification strategy and a sparse draft mechanism ensure that theoretical gains translate into practical efficiency. Experiments across LLMs and benchmarks demonstrate improvements in acceptance rate and speedup over state-of-the-art dynamic tree methods. This framework may provide a template for analyzing stochastic tree structures.

cs.CL↗

arXiv2026

Self-Augmented Diffusion Guidance for Physics-Informed Generation

Diffusion models can be used to generate spatiotemporal signals of physical phenomena, such as time-series images of fluid dynamics. However, a major limitation of standard diffusion models is that they do not incorporate constraints derived from the underlying physical laws. Consequently, generated samples may appear visually plausible while deviating substantially from the true dynamics. In this study, we propose a simple yet effective physics-informed approach based on diffusion guidance with self-generated data augmentation. The proposed method learns the data distribution conditioned on the degree of deviation from the physically correct dynamics and generates samples by explicitly setting the deviation condition to be zero. The method decouples the evaluation of the governing equations from the diffusion model training and sampling processes, avoiding the need to solve the governing equations at every iteration of the denoising process. This design makes the method applicable to problems requiring computationally expensive numerical simulations and enables faster sample generation. Experimental results demonstrate that the proposed model not only significantly reduces the deviations compared with standard diffusion models but also achieves further reductions when combined with existing physics-constrained diffusion methods.

cs.LG↗

arXiv2026

Learning Visually Interpretable Oscillator Networks for Soft Continuum Robots from Video

Learning soft continuum robot (SCR) dynamics from video offers flexibility but existing methods lack interpretability or rely on prior assumptions. Model-based approaches require prior knowledge and manual design. We bridge this gap by introducing: (1) The Attention Broadcast Decoder (ABCD), a plug-and-play module for autoencoder-based latent dynamics learning that generates pixel-accurate attention maps localizing each latent dimension's contribution while filtering static backgrounds, enabling visual interpretability via spatially grounded latents and on-image overlays. (2) Visual Oscillator Networks (VONs), a 2D latent oscillator network coupled to ABCD attention maps for on-image visualization of learned masses, coupling stiffness, and forces, thereby enabling mechanical interpretability. We validate our approach on single- and double-segment SCRs, demonstrating that ABCD-based models significantly improve multi-step prediction accuracy with 5.8x error reduction for Koopman operators and 3.5x for oscillator networks on a two-segment robot. VONs autonomously discover a chain structure of oscillators. This fully data-driven approach yields compact, mechanically interpretable models with potential relevance for future control applications.

cs.RO↗

arXiv2026

Estimating Central, Peripheral, and Temporal Visual Contributions to Human Decision Making in Atari Games

We study how different visual information sources contribute to human decision making in dynamic visual environments. Using Atari-HEAD, a large-scale Atari gameplay dataset with synchronized eye-tracking, we introduce a controlled ablation framework as a means to reverse-engineer the contribution of peripheral visual information, explicit gaze information in the form of gaze maps, and past-state information from human behavior. We train action-prediction networks under six settings that selectively include or exclude these information sources. Across 20 games, peripheral information shows by far the strongest contribution, with median prediction-accuracy drops in the range of 35.27-43.90% when removed. Gaze information yields smaller drops of 2.11-2.76%, while past-state information shows a broader range of 1.52-15.51%, with the upper end likely more informative due to reduced peripheral-information leakage. To complement aggregate accuracies, we cluster states by true-action probabilities assigned by the different model configurations. This analysis identifies coarse behavioral regimes, including focus-dominated, periphery-dominated, and more contextual decision situations. These results suggest that human decision making in Atari depends strongly on information beyond the current focus of gaze, while the proposed framework provides a way to estimate such information-source contributions from behavior.

cs.LG↗

arXiv2026

UniVer: A Unified Perspective for Multi-step and Multi-draft Speculative Decoding

Speculative decoding accelerates Large Language Models via draft-then-verify, where verification can be framed as an Optimal Transport (OT) problem. Existing approaches typically handle multi-draft and multi-step aspects in isolation, applying either flat OT to single-step drafts or per-token rejection sampling to tree-structured candidates. This separation leaves the joint regime (where multi-step dependencies meet multi-draft branching) poorly optimized, as local verification rules fail to exploit the coupling between horizontal and vertical dimensions of candidate trees. In this paper, we propose a unified perspective that casts tree-based verification as a conditional OT problem. Our key insight is that vertical dependencies can be abstracted through prefix acceptance probabilities, which act as dynamic scaling factors to actively guide horizontal draft selection. Based on this principle, we introduce UniVer, a verification algorithm that jointly optimizes across tree levels by composing local optimal transport plans under prefix constraints. We prove that UniVer remains lossless and achieves the optimal acceptance rate under the proposed conditional framework. Extensive experiments across different tasks and models demonstrate that UniVer improves acceptance length by 4.2% to 8.5% over standard recursive rejection sampling without replacement, while maintaining exact distributional alignment with the target model.

cs.CL↗

arXiv2026

Forecast Sports Outcomes under Efficient Market Hypothesis: Theoretical and Experimental Analysis of Odds-Only and Generalised Linear Models

Converting betting odds into accurate outcome probabilities is a fundamental challenge in order to use betting odds as a benchmark for sports forecasting and market efficiency analysis. In this study, we propose two methods to overcome the limitations of existing conversion methods. Firstly, we propose an odds-only method to convert betting odds to probabilities without using historical data for model fitting. While existing odds-only methods, such as Multiplicative, Shin, and Power exist, they do not adjust for biases or relationships we found in our betting odds dataset, which consists of 90014 football matches across five different bookmakers. To overcome these limitations, our proposed Odds-Only-Equal-Profitability-Confidence (OO-EPC) method aligns with the bookmakers' pricing objectives of having equal confidence in profitability for each outcome. We provide empirical evidence from our betting odds dataset that, for the majority of bookmakers, our proposed OO-EPC method outperforms the existing odds-only methods. Beyond controlled experiments, we applied the OO-EPC method under real-world uncertainty by using it for six iterations of an annual basketball outcome forecasting competition. Secondly, we propose a generalised linear model that utilises historical data for model fitting and then converts betting odds to probabilities. Existing generalised linear models attempt to capture relationships that the Efficient Market Hypothesis already captures. To overcome this shortcoming, our proposed Favourite-Longshot-Bias-Adjusted Generalised Linear Model (FL-GLM) fits just one parameter to capture the favourite-longshot bias, providing a more interpretable alternative. We provide empirical evidence from historical football matches where, for all bookmakers, our proposed FL-GLM outperforms the existing multinomial and logistic generalised linear models.

stat.ML↗

arXiv2026

Revealing Human Attention Patterns from Gameplay Analysis for Reinforcement Learning

This study introduces a novel method for revealing human internal attention patterns (decision-relevant attention) from gameplay data alone, leveraging offline attention techniques from reinforcement learning (RL). We propose contextualized, task-relevant (CTR) attention networks, which generate attention maps from both human and RL agent gameplay in Atari environments. To evaluate whether the human CTR maps reveal internal attention patterns, we validate our model by quantitative and qualitative comparison to the agent maps as well as to a temporally integrated overt attention (TIOA) model based on human eye-tracking data. Our results show that human CTR maps are more sparse than the agent ones and align better with the TIOA maps. Following a qualitative visual comparison we conclude that they likely capture patterns of internal attention. As a further application, we use these maps to guide RL agents, finding that human attention-guided agents achieve slightly improved and more stable learning compared to baselines, and significantly outperform TIOA-based agents. This work advances the understanding of human-agent attention differences and provides a new approach for extracting and validating internal attention patterns from behavioral data.

cs.LG↗

arXiv2026

Accurate Open-Loop Control of a Soft Continuum Robot Through Visually Learned Latent Representations

This work addresses open-loop control of a soft continuum robot (SCR) from video-learned latent dynamics. Visual Oscillator Networks (VONs) from previous work are used, that provide mechanistically interpretable 2D oscillator latents through an attention broadcast decoder (ABCD). Open-loop, single-shooting optimal control is performed in latent space to track image-specified waypoints without camera feedback. An interactive SCR live simulator enables design of static, dynamic, and extrapolated targets and maps them to model-specific latent waypoints. On a two-segment pneumatic SCR, Koopman, MLP, and oscillator dynamics, each with and without ABCD, are evaluated on setpoint and dynamic trajectories. ABCD-based models consistently reduce image-space tracking error. The VON and ABCD-based Koopman models attains the lowest MSEs. Using an ablation study, we demonstrate that several architecture choices and training settings contribute to the open-loop control performance. Simulation stress tests further confirm static holding, stable extrapolated equilibria, and plausible relaxation to the rest state. To the best of our knowledge, this is the first demonstration that interpretable, video-learned latent dynamics enable reliable long-horizon open-loop control of an SCR.

cs.RO↗

arXiv2025

A Temporal Difference Method for Stochastic Continuous Dynamics

For continuous systems modeled by dynamical equations such as ODEs and SDEs, Bellman's Principle of Optimality takes the form of the Hamilton-Jacobi-Bellman (HJB) equation, which provides the theoretical target of reinforcement learning (RL). Although recent advances in RL successfully leverage this formulation, the existing methods typically assume the underlying dynamics are known a priori because they need explicit access to the coefficient functions of dynamical equations to update the value function following the HJB equation. We address this inherent limitation of HJB-based RL; we propose a model-free approach still targeting the HJB equation and propose the corresponding temporal difference method. We establish exponential convergence of the idealized continuous-time dynamics and empirically demonstrate its potential advantages over transition-kernel-based formulations. The proposed formulation paves the way toward bridging stochastic control and model-free reinforcement learning.

cs.LG↗

arXiv2021

VIODE: A Simulated Dataset to Address the Challenges of Visual-Inertial Odometry in Dynamic Environments

Dynamic environments such as urban areas are still challenging for popular visual-inertial odometry (VIO) algorithms. Existing datasets typically fail to capture the dynamic nature of these environments, therefore making it difficult to quantitatively evaluate the robustness of existing VIO methods. To address this issue, we propose three contributions: firstly, we provide the VIODE benchmark, a novel dataset recorded from a simulated UAV that navigates in challenging dynamic environments. The unique feature of the VIODE dataset is the systematic introduction of moving objects into the scenes. It includes three environments, each of which is available in four dynamic levels that progressively add moving objects. The dataset contains synchronized stereo images and IMU data, as well as ground-truth trajectories and instance segmentation masks. Secondly, we compare state-of-the-art VIO algorithms on the VIODE dataset and show that they display substantial performance degradation in highly dynamic scenes. Thirdly, we propose a simple extension for visual localization algorithms that relies on semantic information. Our results show that scene semantics are an effective way to mitigate the adverse effects of dynamic objects on VIO algorithms. Finally, we make the VIODE dataset publicly available at https://github.com/kminoda/VIODE.

cs.RO↗

arXiv2020

Companion Unmanned Aerial Vehicles: A Survey

Recent technological advancements in small-scale unmanned aerial vehicles (UAVs) have led to the development of companion UAVs. Similar to conventional companion robots, companion UAVs have the potential to assist us in our daily lives and to help alleviating social loneliness issue. In contrast to ground companion robots, companion UAVs have the capability to fly and possess unique interaction characteristics. Our goals in this work are to have a bird's-eye view of the companion UAV works and to identify lessons learned and guidelines for the design of companion UAVs. We tackle two major challenges towards these goals, where we first use a coordinated way to gather top-quality human-drone interaction (HDI) papers from three sources, and then propose to use a perceptual map of UAVs to summarize current research efforts in HDI. While being simple, the proposed perceptual map can cover the efforts have been made to realize companion UAVs in a comprehensive manner and lead our discussion coherently. We also discuss patterns we noticed in the literature and some lessons learned throughout the review. In addition, we recommend several areas that are worth exploring and suggest a few guidelines to enhance HDI researches with companion UAVs.

cs.RO↗

arXiv2018

Dynamic and Static Topic Model for Analyzing Time-Series Document Collections

For extracting meaningful topics from texts, their structures should be considered properly. In this paper, we aim to analyze structured time-series documents such as a collection of news articles and a series of scientific papers, wherein topics evolve along time depending on multiple topics in the past and are also related to each other at each time. To this end, we propose a dynamic and static topic model, which simultaneously considers the dynamic structures of the temporal topic evolution and the static structures of the topic hierarchy at each time. We show the results of experiments on collections of scientific papers, in which the proposed method outperformed conventional models. Moreover, we show an example of extracted topic structures, which we found helpful for analyzing research activities.

cs.CL↗

arXiv2018

Learning Koopman Invariant Subspaces for Dynamic Mode Decomposition

Spectral decomposition of the Koopman operator is attracting attention as a tool for the analysis of nonlinear dynamical systems. Dynamic mode decomposition is a popular numerical algorithm for Koopman spectral analysis; however, we often need to prepare nonlinear observables manually according to the underlying dynamics, which is not always possible since we may not have any a priori knowledge about them. In this paper, we propose a fully data-driven method for Koopman spectral analysis based on the principle of learning Koopman invariant subspaces from observed data. To this end, we propose minimization of the residual sum of squares of linear least-squares regression to estimate a set of functions that transforms data into a form in which the linear regression fits well. We introduce an implementation with neural networks and evaluate performance empirically using nonlinear dynamical systems and applications.

cs.LG↗

arXiv2017

Recent Developments in Aerial Robotics: A Survey and Prototypes Overview

In recent years, research and development in aerial robotics (i.e., unmanned aerial vehicles, UAVs) has been growing at an unprecedented speed, and there is a need to summarize the background, latest developments, and trends of UAV research. Along with a general overview on the definition, types, categories, and topics of UAV, this work describes a systematic way to identify 1,318 high-quality UAV papers from more than thirty thousand that have been appeared in the top journals and conferences. On top of that, we provide a bird's-eye view of UAV research since 2001 by summarizing various statistical information, such as the year, type, and topic distribution of the UAV papers. We make our survey list public and believe that the list can not only help researchers identify, study, and compare their work, but is also useful for understanding research trends in the field. From our survey results, we find there are many types of UAV, and to the best of our knowledge, no literature has attempted to summarize all types in one place. With our survey list, we explain the types within our survey and outline the recent progress of each. We believe this summary can enhance readers' understanding on the UAVs and inspire researchers to propose new methods and new applications.

cs.RO↗

arXiv2017

Subspace dynamic mode decomposition for stochastic Koopman analysis

The analysis of nonlinear dynamical systems based on the Koopman operator is attracting attention in various applications. Dynamic mode decomposition (DMD) is a data-driven algorithm for Koopman spectral analysis, and several variants with a wide range of applications have been proposed. However, popular implementations of DMD suffer from observation noise on random dynamical systems and generate an inaccurate estimation of the spectra of the stochastic Koopman operator. In this paper, we propose subspace DMD as an algorithm for the Koopman analysis of random dynamical systems with observation noise. Subspace DMD first computes the orthogonal projection of future snapshots to the space of past snapshots and then estimates the spectra of a linear model, and its output converges to the spectra of the stochastic Koopman operator under standard assumptions. We investigate the empirical performance of subspace DMD with several dynamical systems and show its utility for the Koopman analysis of random dynamical systems.

math.DS↗