LoTUS: Large-Scale Machine Unlearning with a Taste of Uncertainty

2025-03-24 · CVPR 2025 1 · Christoforos N. Spartalis, Theodoros Semertzidis, Efstratios Gavves, Petros Daras

We present LoTUS, a novel Machine Unlearning (MU) method that eliminates the influence of training samples from pre-trained models, avoiding retraining from scratch. LoTUS smooths the prediction probabilities of the model up to an information-theoretic bound, mitigating its over-confidence stemming from data memorization. We evaluate LoTUS on Transformer and ResNet18 models against eight baselines across five public datasets. Beyond established MU benchmarks, we evaluate unlearning on ImageNet1k, a large-scale dataset, where retraining is impractical, simulating real-world conditions. Moreover, we introduce the novel Retrain-Free Jensen-Shannon Divergence (RF-JSD) metric to enable evaluation under real-world conditions. The experimental results show that LoTUS outperforms state-of-the-art methods in terms of both efficiency and effectiveness. Code: https://github.com/cspartalis/LoTUS.

📄 PDF Abstract BibTeX arXiv:2503.18314

Code (1)

cspartalis/lotus 공식 구현 pytorch

Tasks

Machine UnlearningMemorization

Methods 이 논문이 사용한 방법론

Attention 설명 없음

Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.

Multi-Head Attention 설명 없음

Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…

Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…

Residual Connection 설명 없음

Adam 설명 없음

Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…

Lotus: learning-based online thermal and latency variation management for two-stage detectors on edge devices

2024-10-01 · Yifan Gong, Yushu Wu, Zheng Zhan, Pu Zhao 외

Two-stage object detectors exhibit high accuracy and precise localization, especially for identifying small objects that are favorable for various edge applications. However, the high computation costs associated with tw…

CPU Deep Reinforcement Learning GPU Management+1

LOTUS: Continual Imitation Learning for Robot Manipulation Through Unsupervised Skill Discovery

2023-11-03 · Weikang Wan, Yifeng Zhu, Rutav Shah, Yuke Zhu

We introduce LOTUS, a continual imitation learning algorithm that empowers a physical robot to continuously and efficiently learn to solve new manipulation tasks throughout its lifespan. The core idea behind LOTUS is con…

Imitation Learning Lifelong learning Robot Manipulation Transfer Learning

Lotus: Efficient LLM Training by Randomized Low-Rank Gradient Projection with Adaptive Subspace Switching

2026-02-01 · Tianhao Miao, Zhongyuan Bao, Lejun Zhang arxiv

Training efficiency in large-scale models is typically assessed through memory consumption, training time, and model performance. Current methods often exhibit trade-offs among these metrics, as optimizing one generally …

Lotus: Creating Short Videos From Long Videos With Abstractive and Extractive Summarization

2025-02-10 · Aadit Barua, Karim Benharrak, Meng Chen, Mina Huh 외

Short-form videos are popular on platforms like TikTok and Instagram as they quickly capture viewers' attention. Many creators repurpose their long-form videos to produce short-form videos, but creators report that plann…

Extractive Summarization Form

AMNESIA: A Large Scale Medical Unlearning Benchmark Suite with Disease-Informed Analysis

2026-05-28 · Saeedeh Davoudi, Reihaneh Iranmanesh, Ophir Frieder, Nazli Goharian arxiv

Medical knowledge is continuously evolving. This creates a need to update or selectively forget information encoded in already-trained medical LLMs. Machine unlearning aims to remove the influence of specific training da…

Clinical Knowledge