Publications

From Inner Randomness to Outer Stability

Nishant Sharma, Ganesh Talluri, Hiroki Naganuma

Under Review, September 2026 (under review)

Which Geometry on Which Layer? A Principled Criterion for Mixed-Optimizer Training

Hiroki Naganuma^† (equal contribution), Laura Gomezjurado Gonzalez^† (equal contribution), Mahdi Ghaznavi^† (equal contribution), Atsushi Nitanda, Ioannis Mitliagkas, Seng Pei Liew, Ryuichiro Hataya

Under Review, May 2026 (under review)

Orth-Dion: Eliminating Geometric Mismatch in Distributed Low-Rank Spectral Optimization

Tatsuhiro Nakamori^† (equal contribution), Laura Gomezjurado Gonzalez^† (equal contribution), Ganesh Talluri^† (equal contribution), Ansh Tiwari, Hideyuki Kawashima, Guillaume Rabusseau, Ioannis Mitliagkas, Hiroki Naganuma

Under Review, May 2026 (under review)

Convergence Bound and Critical Batch Size of Muon Optimizer

Naoki Sato^† (equal contribution), Hiroki Naganuma^† (equal contribution), Hideaki Iiduka

Under Review, July 2025 (under review)

When Does Alignment Help? A Comparative Study of DCCA and Fusion-Based Approaches for Multi-modal Chest X-ray Analysis

Hiroki Naganuma, Akira Uehara, Takeru Sakurai, Yoshiyuki Sankai

SSRN, April 2025

Augmenting NER Datasets with LLMs: Towards Automated and Refined Annotation

Yuji Naraki^† (equal contribution), Ryosuke Yamaki^† (equal contribution), Yoshikazu Ikeda, Takafumi Horie, Kotaro Yoshida, Ryotaro Shimizu, Hiroki Naganuma

arXiv, March 2024

Takeuchi's Information Criteria as Generalization Measures for DNNs Close to NTK Regime

Hiroki Naganuma, Taiji Suzuki, Rio Yokota, Masahiro Nomura, Kohta Ishikawa, Ikuro Sato

arXiv, September 2021

The Geometry of Spectral Gradient Descent: Layerwise Criteria for SignSGD vs SpecSGD

Hiroki Naganuma^† (equal contribution), Laura Gomezjurado Gonzalez^† (equal contribution), Mahdi Ghaznavi^† (equal contribution), Ioannis Mitliagkas

ICLR 2026 Workshop, February 2026

Smoothness-Adaptive Sharpness Aware Minimization for Finding Flatter Minima

Hiroki Naganuma^† (equal contribution), Junhyung Lyle Kim^† (equal contribution), Anastasios Kyrillidis, Ioannis Mitliagkas

ICLR 2024 Workshop, May 2024

workshopposter

Story-to-Images Translation: Leveraging Diffusion Models and Large Language Models for Sequence Image Generation

Haruka Kumagai, Ryosuke Yamaki, Hiroki Naganuma

NarSUM 2023, October 2023

workshoptravel grant

On the Interplay of Curvature, Calibration and Out-of-Distribution Generalization: Insights from SAM and Focal Loss Analyses

Hiroki Naganuma^† (equal contribution), Masanari Kimura^† (equal contribution)

UNCV 2023, October 2023

poster

Necessary and Sufficient Hypothesis of Curvature: Understanding Connection Between Out-of-Distribution Generalization and Calibration

Hiroki Naganuma^† (equal contribution), Masanari Kimura^† (equal contribution)

ICLR 2023 Workshop, May 2023

Towards Understanding the Relationship of Batch Size and Iterations in Deep Learning

Hiroki Naganuma

MLSS 2020, June 2020

On Empirical Analysis of Layer-wise Learning Rate Schedule

Hiroki Naganuma, Rio Yokota

ACML 2019 Workshop, November 2019

A Performance Improvement Approach for Second-Order Optimization in Large Mini-batch Training

Hiroki Naganuma, Rio Yokota

HPML 2019, May 2019

slides

Noise Injection Leads to Better Generalization in Large Mini-Batch Training

Hiroki Naganuma, Rio Yokota

Tokyo Tech-Stony Brook Joint Meeting, May 2019