Research Papers 论文研究 5h ago Updated 45m ago 更新于 45分钟前 49

Anchoring Bias: A Persistent Fairness Backdoor Attack against MLLMs under Continual Learning 锚定偏差:针对持续学习中多模态大语言模型的持久性公平性后门攻击

PFBA (Persistent Fairness Backdoor Attack) is a novel attack that injects group-specific discrimination into Multimodal Large Language Models (MLLMs) that survives continual learning updates The attack uses two mechanisms: Latent Space Fairness Reinforcement (reshapes feature geometry by anchoring privileged-group representations while repelling/clustering targeted-group representations) and Continual Learning Simulation (iteratively optimizes triggers against simulated parameter drift) Naively 提出PFBA(持久公平性后门攻击),首次证明后门诱导的公平性违规可在持续学习过程中长期存活 通过潜空间公平性强化机制重塑特征几何:锚定特权组表示保持模型效用,排斥并聚类目标组表示维持歧视 引入持续学习模拟机制,迭代优化触发器以对抗参数漂移,确保后门在多次更新后仍有效 实验表明PFBA能诱导严重公平性差异,且可规避标准后门防御机制

65
Hot 热度
75
Quality 质量
70
Impact 影响力

Analysis 深度分析

TL;DR

  • PFBA (Persistent Fairness Backdoor Attack) is a novel attack that injects group-specific discrimination into Multimodal Large Language Models (MLLMs) that survives continual learning updates
  • The attack uses two mechanisms: Latent Space Fairness Reinforcement (reshapes feature geometry by anchoring privileged-group representations while repelling/clustering targeted-group representations) and Continual Learning Simulation (iteratively optimizes triggers against simulated parameter drift)
  • Naively implanted backdoors typically degrade under continual learning, but PFBA maintains severe fairness disparities across multiple CL rounds
  • The attack evades standard backdoor defenses while preserving model utility for privileged groups
  • Code and data are publicly available, raising concerns about reproducibility of such attacks

Why It Matters

This research exposes a critical vulnerability in the deployment pipeline of MLLMs: fairness backdoors can persist through continual learning, which is how most production models are updated. For AI practitioners deploying MLLMs in high-stakes domains, this means standard backdoor defenses and periodic model updates may not be sufficient to guarantee fairness, requiring new evaluation and mitigation strategies.

Technical Details

  • Latent Space Fairness Reinforcement: Reshapes the model's deep feature geometry by anchoring privileged-group representations to preserve task utility, while simultaneously repelling and clustering targeted-group representations to sustain discriminatory behavior
  • Continual Learning Simulation: Iteratively optimizes the backdoor trigger against simulated parameter drift, ensuring the attack survives subsequent model updates during continual learning
  • Target: Multimodal Large Language Models (MLLMs) deployed in high-stakes domains where fairness is a safety requirement
  • Key finding: Prior work showed naive backdoors degrade under CL, but PFBA demonstrates that fairness-targeted backdoors can be made persistent through geometric manipulation of latent representations
  • Evaluation: Extensive experiments show PFBA induces severe fairness disparities that persist across continual learning rounds while evading standard backdoor defenses

Industry Insight

  • Organizations deploying MLLMs via continual learning pipelines must treat fairness as a persistent security concern, not a one-time alignment step; periodic updates could inadvertently reinforce or fail to remove embedded biases
  • Standard backdoor detection and mitigation techniques are insufficient against fairness-targeted attacks that exploit the geometric structure of latent representations, necessitating fairness-specific defense mechanisms
  • The public release of code and data lowers the barrier for reproducing such attacks, highlighting an urgent need for the community to develop robust, CL-resilient fairness verification tools before widespread deployment in sensitive domains

TL;DR

  • 提出PFBA(持久公平性后门攻击),首次证明后门诱导的公平性违规可在持续学习过程中长期存活
  • 通过潜空间公平性强化机制重塑特征几何:锚定特权组表示保持模型效用,排斥并聚类目标组表示维持歧视
  • 引入持续学习模拟机制,迭代优化触发器以对抗参数漂移,确保后门在多次更新后仍有效
  • 实验表明PFBA能诱导严重公平性差异,且可规避标准后门防御机制

为什么值得看

本文填补了MLLM安全研究的关键空白,首次系统探讨后门攻击在持续学习场景下的公平性持久性问题。对高风险领域(如医疗、金融、司法)部署MLLMs的从业者具有重要警示意义,揭示了当前持续学习框架在公平性保护方面的潜在漏洞。

技术解析

  • PFBA双机制架构:潜空间公平性强化(Latent Space Fairness Reinforcement)通过几何变换重塑深层特征分布,将特权组表示锚定以保留模型效用,同时将目标组表示排斥并聚类以维持歧视性输出。
  • 持续学习模拟优化:Continual Learning Simulation机制通过模拟参数漂移过程,迭代优化后门触发器,确保攻击在后续模型更新中不被"洗掉"。
  • 实验验证:在多个MLLM基准上验证,PFBA诱导的公平性差异在持续学习多轮后仍保持高水平,且能绕过现有标准后门防御方法。
  • 开源贡献:代码与数据已公开,便于社区复现与后续防御研究。

行业启示

  • 持续学习安全需重新评估:现有后门防御主要针对一次性训练场景,持续学习环境下的公平性安全需要新的评估框架和防御机制。
  • 高风险领域部署需加强监控:在医疗、金融、司法等公平性敏感领域,应建立持续学习过程中的公平性审计机制,防范持久性后门攻击。
  • 防御研究新方向:本文揭示了"公平性后门持久化"这一新型威胁,推动学术界关注持续学习场景下的鲁棒性与公平性联合保障问题。

Disclaimer: The above content is generated by AI and is for reference only. 免责声明:以上内容由 AI 生成,仅供参考。

Security 安全 Multimodal 多模态 LLM 大模型 Research 科学研究 Alignment 对齐