Forced Deferral: Manipulating Routing Decisions in Multimodal LLM Cascades
Quick Answer
This paper shows that The Forced Deferral Attack (FDA) exploits vulnerabilities in multimodal large language model (MLLM) cascades by manipulating the confidence of a weaker model, causing queries to be rerouted to a stronger model.
Quick Take
This adversarial image attack consistently increases strong-model routing across various datasets, outperforming traditional methods like image perturbation. The findings highlight a significant security risk in MLLM systems, where compute allocation can be maliciously influenced without directly affecting answer correctness.
Key Points
- FDA reduces weak model confidence, forcing queries to strong models in MLLM cascades.
- The attack uses a temperature-flattened objective to optimize universal border triggers.
- FDA outperforms image perturbation and prompt injection methods across various benchmarks.
- MLLM cascades expose vulnerabilities that can be exploited to manipulate compute resources.
- This research highlights the need for improved security measures in systems.
Paper Resources
Article Content
From source RSS / original summaryarXiv:2606. 15308v1 Announce Type: new Abstract: While multimodal (MLLMs) have shown strong visual reasoning abilities, serving a large model for every query is computationally expensive. MLLM cascades mitigate this cost by first querying a weak but cheaper model and deferring to a strong model when the weak model's output is unconfident.
However, since the weak model's confidence directly controls compute allocation, these systems expose a new attack surface: an adversary can manipulate confidence so that their queries are consistently deferred to the strong model. Motivated by this vulnerability, we introduce the Forced Deferral Attack (FDA), an adversarial image attack that lowers the weak model's confidence and causes cascades to route queries to the strong model.
FDA learns a universal border trigger by optimizing a temperature-flattened objective. This objective pushes the weak model's token distribution on triggered inputs toward less concentrated targets constructed from its clean responses. Across datasets, model families, and deferral metrics, FDA consistently increases strong-model routing while outperforming image-perturbation and prompt-injection baselines.
These results show that MLLM cascades are vulnerable to attacks that manipulate compute allocation, forcing unintended strong-model usage without directly targeting answer correctness.
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.AI
See more →HOBA: Hierarchical On-Policy Bidding Agents for Adaptive Online Advertising
HOBA (Hierarchical On-policy Bidding Agents) is a novel hierarchical reinforcement learning framework that enhances online advertising bidding systems by improving adaptability and reducing hyperparameter tuning costs. It utilizes a for hyperparameter inference, a SARSA agent for expert model selection, and a dynamic expert pool for bid execution, achieving a +3.6% increase in target cost during large-scale deployment and outperforming state-of-the-art baselines on AuctionNet.