Institution profile

Preferred Networks, Inc.

Industry researchasia · jp
Official website
Research library43linked papers
Opportunities0open roles
Selected work

Representative Papers

Tree-structured Parzen estimator: Understanding its algorithm components and their roles for better empirical performance

Apr 21, 2023arXiv.org

The Tree-structured Parzen Estimator (TPE), a mainstream Bayesian optimization method, lacks a systematic understanding of the functional roles, impact patterns, and synergistic interactions among its key hyperparameters (e.g., γ, n_startup, n_ei_candidates). Method: We conduct comprehensive ablation studies across diverse benchmark functions to empirically dissect these parameters’ behaviors and dependencies. Based on rigorous empirical analysis, we derive an interpretable and robust default configuration strategy. Contribution/Results: Our recommended configuration significantly outperforms the standard TPE implementation and leading baselines—including SMAC and Hyperopt—across heterogeneous benchmarks, improving both optimization efficiency and stability. All experiments are fully reproducible, with source code publicly released.

180 citations11 influentialRead paper

T3lescope: Arbitrary-Resolution High-Fidelity Generative Surface Reconstruction from Images

Oct 02, 2026

This study addresses the inherent trade-off between fixed resolution and spatial coverage in multi-view 3D reconstruction, as well as the challenge of maintaining global geometric consistency in large-scale scenes. We propose a dynamic hierarchical architecture based on diffusion models. During training, only local units are optimized, while inference employs a coarse-to-fine cascading strategy wherein a weight-sharing generator adaptively determines hierarchy levels, scales, and positions. This enables arbitrary-resolution reconstruction without per-scene optimization. The proposed method consistently outperforms existing baselines across indoor to city-scale environments, achieving performance comparable to per-scene optimization approaches while accurately recovering highly reflective and transparent surface structures.

0 citationsRead paper
Recent publications

Latest Papers

T3lescope: Arbitrary-Resolution High-Fidelity Generative Surface Reconstruction from Images

Oct 02, 2026

This study addresses the inherent trade-off between fixed resolution and spatial coverage in multi-view 3D reconstruction, as well as the challenge of maintaining global geometric consistency in large-scale scenes. We propose a dynamic hierarchical architecture based on diffusion models. During training, only local units are optimized, while inference employs a coarse-to-fine cascading strategy wherein a weight-sharing generator adaptively determines hierarchy levels, scales, and positions. This enables arbitrary-resolution reconstruction without per-scene optimization. The proposed method consistently outperforms existing baselines across indoor to city-scale environments, achieving performance comparable to per-scene optimization approaches while accurately recovering highly reflective and transparent surface structures.

0 citationsRead paper

Revisiting Structural Dependency in Autoregressive Multi-Task Table Recognition via Order-Independent Cell-Level Representations

Jun 16, 2026

This work addresses the limitation of autoregressive table recognition methods, whose sequential generation process induces cell representations that are dependent on generation order, thereby compromising global structural consistency. To overcome this, the authors propose a structure refinement module incorporating non-causal attention within a unified multi-task framework, enabling the learning of order-invariant, cell-level feature representations while simultaneously performing table structure prediction, cell localization, and content recognition. By eliminating reliance on autoregressive ordering, the approach substantially enhances both global coherence and parallel inference efficiency. Experimental results on two large-scale datasets demonstrate significant improvements in cell localization and end-to-end recognition accuracy, along with an approximately threefold reduction in overall inference time.

0 citationsRead paper