Exact Unlearning of Finetuning Data via Model Merging at Scale

• URL: http://arxiv.org/abs/2504.04626v1
• Date: Sun, 06 Apr 2025 21:24:29 GMT
• Title: Exact Unlearning of Finetuning Data via Model Merging at Scale
• Authors: Kevin Kuo, Amrith Setlur, Kartik Srinivas, Aditi Raghunathan, Virginia Smith,
• Abstract summary: We propose SIFT-Masks, an exact unlearning method based on model merging.<n>Across four settings where we merge up to 500 models, SIFT-Masks improves accuracy by 5-80% over naive merging.
• Score: 27.352216338702565
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Approximate unlearning has gained popularity as an approach to efficiently update an LLM so that it behaves (roughly) as if it was not trained on a subset of data to begin with. However, existing methods are brittle in practice and can easily be attacked to reveal supposedly unlearned information. To alleviate issues with approximate unlearning, we instead propose SIFT-Masks (SIgn-Fixed Tuning-Masks), an exact unlearning method based on model merging. SIFT-Masks addresses two key limitations of standard model merging: (1) merging a large number of tasks can severely harm utility; and (2) methods that boost utility by sharing extra information across tasks make exact unlearning prohibitively expensive. SIFT-Masks solves these issues by (1) applying local masks to recover task-specific performance; and (2) constraining finetuning to align with a global sign vector as a lightweight approach to determine masks independently before merging. Across four settings where we merge up to 500 models, SIFT-Masks improves accuracy by 5-80% over naive merging and uses up to 250x less compute for exact unlearning compared to other merging baselines.

Related papers

• SHA256 at SemEval-2025 Task 4: Selective Amnesia -- Constrained Unlearning for Large Language Models via Knowledge Isolation [12.838593066237452]
  Large language models (LLMs) memorize frequently sensitive information during training, posing risks when deploying publicly accessible models. This paper presents our solution to SemEval-2025 Task 4 on targeted unlearning, which combines causal mediation analysis with layer-specific optimization.
  arXiv  Detail & Related papers  (2025-04-17T15:05:40Z)
• Multi-Objective Large Language Model Unlearning [3.372396620898397]
  Gradient Ascent (GA) is a proactive way to decrease the prediction probability of the model on the target data. We propose Multi-Objective Large Language Model Unlearning (MOLLM) algorithm to overcome gradient explosion and catastrophic forgetting. Our empirical results verify that MoLLM outperforms the SOTA GA-based LLM unlearning methods in terms of unlearning effect and model utility preservation.
  arXiv  Detail & Related papers  (2024-12-29T09:35:56Z)
• MaskLLM: Learnable Semi-Structured Sparsity for Large Language Models [87.64417894918506]
  This work introduces MaskLLM, a learnable pruning method that establishes Semi-structured (or N:M'') Sparsity in Large Language Models.<n>MaskLLM explicitly models N:M patterns as a learnable distribution through Gumbel Softmax sampling.
  arXiv  Detail & Related papers  (2024-09-26T02:37:41Z)
• Many or Few Samples? Comparing Transfer, Contrastive and Meta-Learning in Encrypted Traffic Classification [68.19713459228369]
  We compare transfer learning, meta-learning and contrastive learning against reference Machine Learning (ML) tree-based and monolithic DL models. We show that (i) using large datasets we can obtain more general representations, (ii) contrastive learning is the best methodology. While ML tree-based cannot handle large tasks but fits well small tasks, by means of reusing learned representations, DL methods are reaching tree-based models performance also for small tasks.
  arXiv  Detail & Related papers  (2023-05-21T11:20:49Z)
• Masked Autoencoding for Scalable and Generalizable Decision Making [93.84855114717062]
  MaskDP is a simple and scalable self-supervised pretraining method for reinforcement learning and behavioral cloning. We find that a MaskDP model gains the capability of zero-shot transfer to new BC tasks, such as single and multiple goal reaching.
  arXiv  Detail & Related papers  (2022-11-23T07:04:41Z)
• Bi-level Alignment for Cross-Domain Crowd Counting [113.78303285148041]
  Current methods rely on external data for training an auxiliary task or apply an expensive coarse-to-fine estimation. We develop a new adversarial learning based method, which is simple and efficient to apply. We evaluate our approach on five real-world crowd counting benchmarks, where we outperform existing approaches by a large margin.
  arXiv  Detail & Related papers  (2022-05-12T02:23:25Z)
• Fast Yet Effective Machine Unlearning [6.884272840652062]
  We introduce a novel machine unlearning framework with error-maximizing noise generation and impair-repair based weight manipulation. We show excellent unlearning while substantially retaining the overall model accuracy. This work is an important step towards fast and easy implementation of unlearning in deep networks.
  arXiv  Detail & Related papers  (2021-11-17T07:29:24Z)
• Meta-Generating Deep Attentive Metric for Few-shot Classification [53.07108067253006]
  We present a novel deep metric meta-generation method to generate a specific metric for a new few-shot learning task. In this study, we structure the metric using a three-layer deep attentive network that is flexible enough to produce a discriminative metric for each task. We gain surprisingly obvious performance improvement over state-of-the-art competitors, especially in the challenging cases.
  arXiv  Detail & Related papers  (2020-12-03T02:07:43Z)
• KSM: Fast Multiple Task Adaption via Kernel-wise Soft Mask Learning [49.77278179376902]
  Deep Neural Networks (DNN) could forget the knowledge about earlier tasks when learning new tasks, and this is known as textitcatastrophic forgetting. Recent continual learning methods are capable of alleviating the catastrophic problem on toy-sized datasets. We propose a new training method called textit- Kernel-wise Soft Mask (KSM), which learns a kernel-wise hybrid binary and real-value soft mask for each task.
  arXiv  Detail & Related papers  (2020-09-11T21:48:39Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.