7 papers
Sparse-Autoencoder-Guided Internal Representation Unlearning for Large Language Models
Tomoya Yamashita, Akira Ito, Yuuki Yamanaka +3
As large language models (LLMs) are increasingly deployed across various applications, privacy and copyright concerns have heightened the need for more effective LLM unlearning tec…
Concept Unlearning in Large Language Models via Self-Constructed Knowledge Triplets
Tomoya Yamashita, Yuuki Yamanaka, Masanori Yamada +3
Machine Unlearning (MU) has recently attracted considerable attention as a solution to privacy and copyright issues in large language models (LLMs). Existing MU methods aim to remo…
Positive-Unlabeled Diffusion Models for Preventing Sensitive Data Generation
Hiroshi Takahashi, Tomoharu Iwata, Atsutoshi Kumagai +2
Diffusion models are powerful generative models but often generate sensitive data that are unwanted by users, mainly because the unlabeled training data frequently contain such sen…
Dialogue System of Team NTT-EASE for DRC2023
Yuki Kubo, Tomoya Yamashita, Masanori Yamada
We developed a dialogue system as a team NTT-EASE in the Dialogue Robot Competition 2023 (DRC2023). We introduce a dialogue system (EASE-DRCBot) constructed for DRC2023. EASE-DRCBo…
One-Shot Machine Unlearning with Mnemonic Code
Tomoya Yamashita, Masanori Yamada, Takashi Shibata
Ethical and privacy issues inherent in artificial intelligence (AI) applications have been a growing concern with the rapid spread of deep learning. Machine unlearning (MU) is the…
Toward Data Efficient Model Merging between Different Datasets without Performance Degradation
Masanori Yamada, Tomoya Yamashita, Shin'ya Yamaguchi +1
Model merging is attracting attention as a novel method for creating a new model by combining the weights of different trained models. While previous studies reported that model me…
ARDIR: Improving Robustness using Knowledge Distillation of Internal Representation
Tomokatsu Takahashi, Masanori Yamada, Yuuki Yamanaka +1
Adversarial training is the most promising method for learning robust models against adversarial examples. A recent study has shown that knowledge distillation between the same arc…