Skip to content

Anchoring Bias: A Persistent Fairness Backdoor Attack against MLLMs under Continual Learning

Aug 2026 · ACM International Conference on Information and Knowledge Management (CIKM) · 0 citations
Computer Science

Abstract

Multimodal Large Language Models (MLLMs) are increasingly deployed in high-stakes domains where fairness is a critical safety requirement. In practice, these models are continually updated through continual learning (CL) to adapt to evolving tasks and data distributions. Prior work has shown that backdoor attacks can manipulate MLLM responses through hidden triggers, but naively implanted backdoors degrade as models undergo subsequent updates of CL. Although fairness has emerged as a central concern for MLLM deployment, whether backdoor-induced fairness violations can survive CL remains unexplored, leaving two critical questions unanswered: (1) whether a backdoor can reliably induce fairness violations in MLLMs, and (2) whether such fairness-targeted backdoors can persist through continual learning. We bridge this gap by proposing Persistent Fairness Backdoor Attack (PFBA) to inject persistent and group-specific discrimination into MLLMs. Specifically, PFBA achieves this through two novel mechanisms. The Latent Space Fairness Reinforcement reshapes the model's deep feature geometry by anchoring privileged-group representations to preserve utility while repelling and clustering targeted-group representations to sustain discrimination, and the Continual Learning Simulation iteratively optimizes the trigger against simulated parameter drift to ensure backdoor persistence across future updates. Extensive experiments demonstrate that PFBA induces severe fairness disparities that persist across continual learning rounds, evading standard backdoor defenses. The data and code are publicly available at https://github.com/lyygua/PFBA.

View source

Similar papers

#machine learning Review Open access Oct 2014

Software development in startup companies: A systematic mapping study

The results indicate that software engineering work practices are chosen opportunistically, adapted and configured to provide value under the constrains imposed by the startup context.

Nicolò Paternoster, Carmine Giardino, M. Unterkalmsteiner et al. · 394 citations · ⚡54
#machine learning Review Open access Jun 2014

Why Early-Stage Software Startups Fail: A Behavioral Framework

This state-of-practice investigation was performed using a literature review followed by a multiple-case study approach and presents how inconsistency between managerial strategies and execution can lead to failure by means of a behavioral framework.

Carmine Giardino, Xiaofeng Wang, P. Abrahamsson · 175 citations · ⚡19
#machine learning Review Open access Oct 2016

“Failures” to be celebrated: an analysis of major pivots of software startups

This study conducts a case survey study based on the secondary data of the major pivots happened in 49 software startups, and demonstrates that customer need pivot is the most common among all pivot types.

Sohaib Shahid Bajwa, Xiaofeng Wang, Anh Nguyen-Duc et al. · 127 citations · ⚡15
#machine learning Review Open access May 2016

Key Challenges in Software Startups Across Life Cycle Stages

It is found that what perceived as biggest challenges by software startups do vary across different life cycle stages, even though its significance decreases when the learning focuses of the startups move from problem to solution and their products mature.

Xiaofeng Wang, Henry Edison, Sohaib Shahid Bajwa et al. · 62 citations · ⚡6

Related blog posts

MIT News · Artificial Intelligence Nov 17, 2026

Learning the language of machine learning

By studying the computational foundations of language learning, Associate Professor Jacob Andreas builds intelligent systems that can communicate effectively with humans.

Microsoft Research Blog Nov 16, 2026

TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient

AI coding agents are increasingly help developers fix bugs and build features largely on their own, but every interaction with their models consumes tokens. In each session, an agent sends instructions, tool descriptions, and session history to its model. Most efforts to reduce this cost focus on individual tasks. Yet developers return to the same […] The post TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient appeared first on Microsoft Research.

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.