paper-with-me

홈 › Papers

Harnessing Hyperbolic Geometry for Harmful Prompt Detection and Sanitization

2026-04-07 · Igor Maljkovic, Maria Rosaria Briglia, Iacopo Masi, Antonio Emanuele Cinà, Fabio Roli arxiv

Vision-Language Models (VLMs) have become essential for tasks such as image synthesis, captioning, and retrieval by aligning textual and visual information in a shared embedding space. Yet, this flexibility also makes them vulnerable to malicious prompts designed to produce unsafe content, raising critical safety concerns. Existing defenses either rely on blacklist filters, which are easily circumvented, or on heavy classifier-based systems, both of which are costly and fragile under embedding-level attacks. We address these challenges with two complementary components: Hyperbolic Prompt Espial (HyPE) and Hyperbolic Prompt Sanitization (HyPS). HyPE is a lightweight anomaly detector that leverages the structured geometry of hyperbolic space to model benign prompts and detect harmful ones as outliers. HyPS builds on this detection by applying explainable attribution methods to identify and selectively modify harmful words, neutralizing unsafe intent while preserving the original semantics of user prompts. Through extensive experiments across multiple datasets and adversarial scenarios, we prove that our framework consistently outperforms prior defenses in both detection accuracy and robustness. Together, HyPE and HyPS offer an efficient, interpretable, and resilient approach to safeguarding VLMs against malicious prompt misuse.

📄 PDF Abstract BibTeX arXiv:2604.06285

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

The Geometry of Harmful Intent: Training-Free Anomaly Detection via Angular Deviation in LLM Residual Streams

2026-03-28 · Isaac Llorente-Saguer arxiv

We present LatentBiopsy, a training-free method for detecting harmful prompts by analysing the geometry of residual-stream activations in large language models. Given 200 safe normative prompts, LatentBiopsy computes the…

Anomaly Detection

Hyperbolic Metric Learning for Visual Outlier Detection

2024-03-22 · Alvaro Gonzalez-Jimenez, Simone Lionetti, Dena Bazazian, Philippe Gottfrois 외

Out-Of-Distribution (OOD) detection is critical to deploy deep learning models in safety-critical applications. However, the inherent hierarchical concept structure of visual data, which is instrumental to OOD detection,…

Metric LearningOutlier DetectionOut of Distribution (OOD) Detection

Minimizing Hyperbolic Embedding Distortion with LLM-Guided Hierarchy Restructuring

2025-11-16 · Melika Ayoughi, Pascal Mettes, Paul Groth arxiv

Hyperbolic geometry is an effective geometry for embedding hierarchical data structures. Hyperbolic learning has therefore become increasingly prominent in machine learning applications where data is hierarchically organ…

Recommendation SystemsKnowledge Graphs

PUZZLED: Jailbreaking LLMs through Word-Based Puzzles

2025-08-02 · Yelim Ahn, Jaejin Lee arxiv

As large language models (LLMs) are increasingly deployed across diverse domains, ensuring their safety has become a critical concern. In response, studies on jailbreak attacks have been actively growing. Existing approa…

Prompt Engineering

HyperbolicDiffusion: Sharp & Scalable Tiled Generation on the Hyperbolic Plane

2026-08-04 · Hugo Caselles-Dupré arxiv

Planar tiled diffusion denoises overlapping windows of one rectangular canvas. The hyperbolic plane has no such canvas, and its area grows exponentially with radius. We introduce HyperbolicDiffusion, a training-free meth…