41 citations · 109 across the 19 of their papers we have counts for
21 papers
PrivCert: Certifying Statement Support under Differential Privacy
Tsubasa Takahashi, Takumi Hiraoka
Differentially private (DP) text generation can protect individual records, but privacy alone does not specify what evidence a released statement carries about the underlying data.…
One-D-Piece: Image Tokenizer Meets Quality-Controllable Compression
Keita Miwa, Kento Sasaki, Hidehisa Arai +2
Current image tokenization methods require a large number of tokens to capture the information contained within images. Although the amount of information varies across images, mos…
ACT-Bench: Towards Action Controllable World Models for Autonomous Driving
Hidehisa Arai, Keishi Ishihara, Tsubasa Takahashi +1
World models have emerged as promising neural simulators for autonomous driving, with the potential to supplement scarce real-world data and enable closed-loop evaluations. However…
Self-Preference Bias in LLM-as-a-Judge
Koki Wataoka, Tsubasa Takahashi, Ryokan Ri
Automated evaluation leveraging large language models (LLMs), commonly referred to as LLM evaluators or LLM-as-a-judge, has been widely used in measuring the performance of dialogu…
MergePrint: Merge-Resistant Fingerprints for Robust Black-box Ownership Verification of Large Language Models
Shojiro Yamabe, Futa Waseda, Tsubasa Takahashi +1
Protecting the intellectual property of Large Language Models (LLMs) has become increasingly critical due to the high cost of training. Model merging, which integrates multiple exp…
Federated Graph Analytics with Differential Privacy
Shang Liu, Yang Cao, Takao Murakami +5
Collaborative graph analysis across multiple institutions is becoming increasingly popular. Realistic examples include social network analysis across various social platforms, fina…
Watermark-embedded Adversarial Examples for Copyright Protection against Diffusion Models
Peifei Zhu, Tsubasa Takahashi, Hirokatsu Kataoka
Diffusion Models (DMs) have shown remarkable capabilities in various image-generation tasks. However, there are growing concerns that DMs could be used to imitate unauthorized crea…
Scaling Private Deep Learning with Low-Rank and Sparse Gradients
Ryuichi Ito, Seng Pei Liew, Tsubasa Takahashi +2
Applying Differentially Private Stochastic Gradient Descent (DPSGD) to training modern, large-scale neural networks such as transformer-based models is a challenging task, as the m…
Shuffle Gaussian Mechanism for Differential Privacy
Seng Pei Liew, Tsubasa Takahashi
We study Gaussian mechanism in the shuffle model of differential privacy (DP). Particularly, we characterize the mechanism's Rényi differential privacy (RDP), showing that it is of…
Measuring Lower Bounds of Local Differential Privacy via Adversary Instantiations in Federated Learning
Marin Matsumoto, Tsubasa Takahashi, Seng Pei Liew +1
Local differential privacy (LDP) gives a strong privacy guarantee to be used in a distributed setting like federated learning (FL). LDP mechanisms in FL protect a client's gradient…
Privacy Amplification via Shuffled Check-Ins
Seng Pei Liew, Satoshi Hasegawa, Tsubasa Takahashi
We study a protocol for distributed computation called shuffled check-in, which achieves strong privacy guarantees without requiring any further trust assumptions beyond a trusted…
Network Shuffling: Privacy Amplification via Random Walks
Seng Pei Liew, Tsubasa Takahashi, Shun Takagi +3
Recently, it is shown that shuffling can amplify the central differential privacy guarantees of data randomized with local differential privacy. Within this setup, a centralized, t…
HDPView: Differentially Private Materialized View for Exploring High Dimensional Relational Data
Fumiyuki Kato, Tsubasa Takahashi, Shun Takagi +3
How can we explore the unknown properties of high-dimensional sensitive relational data while preserving privacy? We study how to construct an explorable privacy-preserving materia…
Construction of Differentially Private Summaries over Fully Homomorphic Encryption
S. Ushiyama, T. Takahashi, M. Kudo +1
Cloud computing has garnered attention as a platform of query processing systems. However, data privacy leakage is a critical problem. Chowdhury et al. proposed Crypt(epsilon), whi…
PEARL: Data Synthesis via Private Embeddings and Adversarial Reconstruction Learning
Seng Pei Liew, Tsubasa Takahashi, Michihiko Ueno
We propose a new framework of synthesizing data using deep generative models in a differentially private manner. Within our framework, sensitive data are sanitized with rigorous pr…
FaceLeaks: Inference Attacks against Transfer Learning Models via Black-box Queries
Seng Pei Liew, Tsubasa Takahashi
Transfer learning is a useful machine learning framework that allows one to build task-specific models (student models) without significantly incurring training costs using a singl…
Differentially Private Variational Autoencoders with Term-wise Gradient Aggregation
Tsubasa Takahashi, Shun Takagi, Hajime Ono +1
This paper studies how to learn variational autoencoders with a variety of divergences under differential privacy constraints. We often build a VAE with an appropriate prior distri…
P3GM: Private High-Dimensional Data Release via Privacy Preserving Phased Generative Model
Shun Takagi, Tsubasa Takahashi, Yang Cao +1
How can we release a massive volume of sensitive data while mitigating privacy risks? Privacy-preserving data synthesis enables the data holder to outsource analytical tasks to an…
Indirect Adversarial Attacks via Poisoning Neighbors for Graph Convolutional Networks
Tsubasa Takahashi
Graph convolutional neural networks, which learn aggregations over neighbor nodes, have achieved great performance in node classification tasks. However, recent studies reported th…
Locally Private Distributed Reinforcement Learning
Hajime Ono, Tsubasa Takahashi
We study locally differentially private algorithms for reinforcement learning to obtain a robust policy that performs well across distributed private environments. Our algorithm pr…
Lightweight Lipschitz Margin Training for Certified Defense against Adversarial Examples
Hajime Ono, Tsubasa Takahashi, Kazuya Kakizaki
How can we make machine learning provably robust against adversarial examples in a scalable way? Since certified defense methods, which ensure -robust, consume huge resources, t…