Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning

• URL: http://arxiv.org/abs/2506.03850v1
• Date: Wed, 04 Jun 2025 11:33:36 GMT
• Title: Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning
• Authors: Liang Chen, Xueting Han, Li Shen, Jing Bai, Kam-Fai Wong,
• Abstract summary: Vulnerability-Aware Alignment estimates data vulnerability, partitions data into "vulnerable" and "invulnerable" groups, and encourages balanced learning.<n>VAA significantly reduces harmful scores while preserving downstream task performance, outperforming state-of-the-art baselines.
• Score: 22.13346397293792
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Harmful fine-tuning (HFT), performed directly on open-source LLMs or through Fine-tuning-as-a-Service, breaks safety alignment and poses significant threats. Existing methods aim to mitigate HFT risks by learning robust representation on alignment data or making harmful data unlearnable, but they treat each data sample equally, leaving data vulnerability patterns understudied. In this work, we reveal that certain subsets of alignment data are consistently more prone to forgetting during HFT across different fine-tuning tasks. Inspired by these findings, we propose Vulnerability-Aware Alignment (VAA), which estimates data vulnerability, partitions data into "vulnerable" and "invulnerable" groups, and encourages balanced learning using a group distributionally robust optimization (Group DRO) framework. Specifically, VAA learns an adversarial sampler that samples examples from the currently underperforming group and then applies group-dependent adversarial perturbations to the data during training, aiming to encourage a balanced learning process across groups. Experiments across four fine-tuning tasks demonstrate that VAA significantly reduces harmful scores while preserving downstream task performance, outperforming state-of-the-art baselines.

Related papers

• FairSAM: Fair Classification on Corrupted Data Through Sharpness-Aware Minimization [12.178322948983263]
  Image classification models trained on clean data often suffer from significant performance degradation when exposed to testing corrupted data.<n>This degradation not only impacts overall performance but also disproportionately affects various demographic subgroups, raising critical algorithmic bias concerns.<n>Existing fairness-aware machine learning methods aim to reduce performance disparities but hardly maintain robust and equitable accuracy when faced with data corruption.<n>We propose textbfFairSAM, a new framework that integrates underlineFairness-oriented strategies into underlineSAM to deliver equalized performance across demographic groups under corrupted conditions.
  arXiv  Detail & Related papers  (2025-03-29T01:51:59Z)
• Trained Models Tell Us How to Make Them Robust to Spurious Correlation without Group Annotation [3.894771553698554]
  Empirical Risk Minimization (ERM) models tend to rely on attributes that have high spurious correlation with the target. This can degrade the performance on underrepresented (or'minority') groups that lack these attributes. We propose Environment-based Validation and Loss-based Sampling (EVaLS) to enhance robustness to spurious correlation.
  arXiv  Detail & Related papers  (2024-10-07T08:17:44Z)
• Uncertainty Aware Learning for Language Model Alignment [97.36361196793929]
  We propose uncertainty-aware learning (UAL) to improve the model alignment of different task scenarios. We implement UAL in a simple fashion -- adaptively setting the label smoothing value of training according to the uncertainty of individual samples. Experiments on widely used benchmarks demonstrate that our UAL significantly and consistently outperforms standard supervised fine-tuning.
  arXiv  Detail & Related papers  (2024-06-07T11:37:45Z)
• Noisy Correspondence Learning with Self-Reinforcing Errors Mitigation [63.180725016463974]
  Cross-modal retrieval relies on well-matched large-scale datasets that are laborious in practice. We introduce a novel noisy correspondence learning framework, namely textbfSelf-textbfReinforcing textbfErrors textbfMitigation (SREM)
  arXiv  Detail & Related papers  (2023-12-27T09:03:43Z)
• MAPS: A Noise-Robust Progressive Learning Approach for Source-Free Domain Adaptive Keypoint Detection [76.97324120775475]
  Cross-domain keypoint detection methods always require accessing the source data during adaptation. This paper considers source-free domain adaptive keypoint detection, where only the well-trained source model is provided to the target domain.
  arXiv  Detail & Related papers  (2023-02-09T12:06:08Z)
• CAFA: Class-Aware Feature Alignment for Test-Time Adaptation [50.26963784271912]
  Test-time adaptation (TTA) aims to address this challenge by adapting a model to unlabeled data at test time. We propose a simple yet effective feature alignment loss, termed as Class-Aware Feature Alignment (CAFA), which simultaneously encourages a model to learn target representations in a class-discriminative manner.
  arXiv  Detail & Related papers  (2022-06-01T03:02:07Z)
• Examining and Combating Spurious Features under Distribution Shift [94.31956965507085]
  We define and analyze robust and spurious representations using the information-theoretic concept of minimal sufficient statistics. We prove that even when there is only bias of the input distribution, models can still pick up spurious features from their training data. Inspired by our analysis, we demonstrate that group DRO can fail when groups do not directly account for various spurious correlations.
  arXiv  Detail & Related papers  (2021-06-14T05:39:09Z)
• Negative Data Augmentation [127.28042046152954]
  We show that negative data augmentation samples provide information on the support of the data distribution. We introduce a new GAN training objective where we use NDA as an additional source of synthetic data for the discriminator. Empirically, models trained with our method achieve improved conditional/unconditional image generation along with improved anomaly detection capabilities.
  arXiv  Detail & Related papers  (2021-02-09T20:28:35Z)
• Provably Efficient Causal Reinforcement Learning with Confounded Observational Data [135.64775986546505]
  We study how to incorporate the dataset (observational data) collected offline, which is often abundantly available in practice, to improve the sample efficiency in the online setting. We propose the deconfounded optimistic value iteration (DOVI) algorithm, which incorporates the confounded observational data in a provably efficient manner.
  arXiv  Detail & Related papers  (2020-06-22T14:49:33Z)

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.