The Mirror Design Pattern: Strict Data Geometry over Model Scale for Prompt Injection Detection

• URL: http://arxiv.org/abs/2603.11875v1
• Date: Thu, 12 Mar 2026 12:50:57 GMT
• Title: The Mirror Design Pattern: Strict Data Geometry over Model Scale for Prompt Injection Detection
• Abstract summary: We introduce Mirror, a data-curation design pattern that organizes prompt injection corpora into matched positive and negative cells.<n>We show that for L1 prompt injection screening, strict data geometry can matter more than model scale.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Prompt injection defenses are often framed as semantic understanding problems and delegated to increasingly large neural detectors. For the first screening layer, however, the requirements are different: the detector runs on every request and therefore must be fast, deterministic, non-promptable, and auditable. We introduce Mirror, a data-curation design pattern that organizes prompt injection corpora into matched positive and negative cells so that a classifier learns control-plane attack mechanics rather than incidental corpus shortcuts. Using 5,000 strictly curated open-source samples -- the largest corpus supportable under our public-data validity contract -- we define a 32-cell mirror topology, fill 31 of those cells with public data, train a sparse character n-gram linear SVM, compile its weights into a static Rust artifact, and obtain 95.97\% recall and 92.07\% F1 on a 524-case holdout at sub-millisecond latency with no external model runtime dependencies. On the same holdout, our next line of defense, a 22-million-parameter Prompt Guard~2 model reaches 44.35\% recall and 59.14\% F1 at 49\,ms median and 324\,ms p95 latency. Linear models still leave residual semantic ambiguities such as use-versus-mention for later pipeline layers, but within that scope our results show that for L1 prompt injection screening, strict data geometry can matter more than model scale.

Related papers

• TriCalRAG: A Three-Strategy, Retrieval-Augmented Benchmark for On-Premise LLM-Based Root Cause Analysis in AIOps [1.8136615181861766]
  Cloud-hosted large language models (LLMs) are increasingly used for root cause analysis (RCA) in AIOps pipelines.<n>They introduce data privacy risk, network latency, and per-query cost that scale poorly with production log volumes.<n>We presentRAG, a benchmark evaluating open-weights served locally via vLLM against a classical LSTM-based log anomaly detector.
  arXiv  Detail & Related papers  (2026-09-13T19:52:46Z)
• Constrained Decoding for Diffusion Language Models via Efficient Inference over Finite Automata [57.27430779838529]
  We present an exact and tractable algorithm for sampling from the constrained mean-field posterior under any constraint expressible as a finite automaton.<n>The approach guarantees constraint satisfaction by construction, supports both greedy and sampling-based decoding, and is compatible with parallel and block-wise decoding.<n> Empirical evaluations on Dream-7B and LLaDA-8 show substantial accuracy gains across various tasks.
  arXiv  Detail & Related papers  (2026-07-08T05:48:57Z)
• Towards Whole Hand and Wrist Kinematic Tracking with a Wearable A-Mode Ultrasound Probe [13.901959380411043]
  A-mode ultrasound (US) has emerged as a promising modality for hand and wrist motion tracking.<n>We propose a framework for robust whole-hand and wrist kinematic tracking via wearable A-mode US using the WULPUS platform.
  arXiv  Detail & Related papers  (2026-06-21T04:52:02Z)
• TAME: Test-Time Adversarial Prompt Tuning via Mixture-of-Experts for Vision-Language Models [76.92890872338235]
  Large-scale pre-trained Vision-Language models (VLMs) exhibit strong zero-shot generalization, yet are highly vulnerable to imperceptible adversarial perturbations.<n>To enhance robustness without requiring downstream task-specific retraining, we propose TAME, a novel test-time defense.
  arXiv  Detail & Related papers  (2026-05-17T18:07:08Z)
• EntropyScan: Towards Model-level Backdoor Detection in LVLMs via Visual Attention Entropy [66.59724477993339]
  Large Vision-Language Models (LVLMs) have demonstrated remarkable capabilities across various tasks, yet they remain vulnerable to backdoor attacks.<n>We propose EntropyScan, a lightweight and trigger-agnostic method for model-level backdoor detection in LVLMs.<n>Our code will be publicly available soon.
  arXiv  Detail & Related papers  (2026-05-15T08:01:32Z)
• Anomaly Detection in IEC-61850 GOOSE Networks: Evaluating Unsupervised and Temporal Learning for Real-Time Intrusion Detection [0.0]
  The IEC-61850 GOOSE protocol underpins time-critical communication in modern digital substations.<n>Intrusion detection in this setting is challenging due to strict latency constraints (sub-4ms) and limited availability of labeled attack data.<n>This paper evaluates whether unsupervised temporal modeling can provide effective and deployable anomaly detection for GOOSE networks.
  arXiv  Detail & Related papers  (2026-04-14T20:13:55Z)
• Detect Anything in Real Time: From Single-Prompt Segmentation to Multi-Class Detection [0.0]
  We present a training-free framework that converts SAM3 into a real-time multi-class detector by exploiting a structural invariant.<n>On val 2017 (5,000 images, 80 classes), DART achieves 55.8 AP at 15.8 FPS (4 classes, 1008x1008) on a single GTX 4080.<n>For extreme latency targets, adapter distillation with a frozen encoder-decoder achieves 38.7 AP with a 13.9 ms backbone.
  arXiv  Detail & Related papers  (2026-03-12T02:05:11Z)
• ReasoningBomb: A Stealthy Denial-of-Service Attack by Inducing Pathologically Long Reasoning in Large Reasoning Models [67.15960154375131]
  Large reasoning models (LRMs) extend large language models with explicit multi-step reasoning traces.<n>This capability introduces a new class of prompt-induced inference-time denial-of-service (PI-DoS) attacks that exploit the high computational cost of reasoning.<n>We present ReasoningBomb, a reinforcement-learning-based PI-DoS framework that is guided by a constant-time surrogate reward.
  arXiv  Detail & Related papers  (2026-01-29T18:53:01Z)
• LiQSS: Post-Transformer Linear Quantum-Inspired State-Space Tensor Networks for Real-Time 6G [85.58816960936069]
  Proactive and agentic control in Sixth-Generation (6G) Open Radio Access Networks (O-RAN) requires control-grade prediction under stringent Near-Time (Near-RT) latency and computational constraints.<n>This paper investigates a post-Transformer paradigm for efficient radio telemetry forecasting.<n>We propose a quantum-inspired state-space tensor network that replaces self-attention with stable structured state-space dynamics kernels.
  arXiv  Detail & Related papers  (2026-01-18T12:08:38Z)
• Efficient Jailbreak Mitigation Using Semantic Linear Classification in a Multi-Staged Pipeline [1.2802720336459552]
  Prompt injection and jailbreaking attacks pose persistent security challenges to large language model (LLM)-based systems.<n>We present an efficient and systematically evaluated defense architecture that mitigates these threats through a lightweight, multi-stage pipeline.
  arXiv  Detail & Related papers  (2025-12-22T04:00:35Z)
• HuggingR$^{4}$: A Progressive Reasoning Framework for Discovering Optimal Model Companions [50.61510609116118]
  HuggingR$4$ is a novel framework that combines Reasoning, Retrieval, Refinement, and Reflection to efficiently select models.<n>It attains a workability rate of 92.03% and a reasonability rate of 82.46%, surpassing existing method by 26.51% and 33.25% respectively.
  arXiv  Detail & Related papers  (2025-11-24T03:13:45Z)
• Memory-SAM: Human-Prompt-Free Tongue Segmentation via Retrieval-to-Prompt [4.712714373340359]
  Memory-SAM is a training-free, human-prompt-free pipeline.<n>It automatically generates effective prompts from a small memory of prior cases.<n>Results indicate that retrieval-to-prompt enables data-efficient, robust segmentation of irregular boundaries in tongue imaging.
  arXiv  Detail & Related papers  (2025-10-17T17:42:28Z)
• PLEIADES: Building Temporal Kernels with Orthogonal Polynomials [6.2921559513184695]
  We introduce a class of neural networks named PLEIA (Lynomial Expansion In Adaptive Distributed Event Systems), which contains temporal kernels generated from basis functions.<n>We focus on interfacing these networks with event-based data to perform online classification and detection with low latency.
  arXiv  Detail & Related papers  (2024-05-20T17:06:24Z)
• Tracking Meets LoRA: Faster Training, Larger Model, Stronger Performance [87.19164603145056]
  We propose LoRAT, a method that unveils the power of large ViT model for tracking within laboratory-level resources. The essence of our work lies in adapting LoRA, a technique that fine-tunes a small subset of model parameters without adding inference latency. We design an anchor-free head solely based on to adapt PETR, enabling better performance with less computational overhead.
  arXiv  Detail & Related papers  (2024-03-08T11:41:48Z)
• Cryptanalytic Extraction of Neural Network Models [56.738871473622865]
  We introduce a differential attack that can efficiently steal the parameters of the remote model up to floating point precision. Our attack relies on the fact that ReLU neural networks are piecewise linear functions. We extract models that are 220 times more precise and require 100x fewer queries than prior work.
  arXiv  Detail & Related papers  (2020-03-10T17:57:14Z)

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.