paper-with-me

Papers

SPIN: Self-Supervised Prompt INjection

2024-10-17 · Leon Zhou, Junfeng Yang, Chengzhi Mao

Large Language Models (LLMs) are increasingly used in a variety of important applications, yet their safety and reliability remain as major concerns. Various adversarial and jailbreak attacks have been proposed to bypass the safety alignment and cause the model to produce harmful responses. We introduce Self-supervised Prompt INjection (SPIN) which can detect and reverse these various attacks on LLMs. As our self-supervised prompt defense is done at inference-time, it is also compatible with existing alignment and adds an additional layer of safety for defense. Our benchmarks demonstrate that our system can reduce the attack success rate by up to 87.9%, while maintaining the performance on benign user requests. In addition, we discuss the situation of an adaptive attacker and show that our method is still resilient against attackers who are aware of our defense.

📄 PDF Abstract BibTeX arXiv:2410.13236

Code (0)

등록된 구현이 없습니다.

Tasks

Safety Alignment

Methods 이 논문이 사용한 방법론

AWARE We propose to theoretically and empirically examine the effect of incorporating weighting schemes into walk-aggregating GNNs. To this end, we propose a simple, interpretable, and…

Similar Papers 제목 키워드 기반

Self-interpreting Adversarial Images

2024-07-12 · Tingwei Zhang, Collin Zhang, John X. Morris, Eugene Bagdasarian 외

We introduce a new type of indirect, cross-modal injection attacks against visual language models that enable creation of self-interpreting images. These images contain hidden "meta-instructions" that control how models …

Misinformation

CPS4: Class Prompt driven Semi-Supervised Spine Segmentation with Class-specific Consistency Constraint

2026-06-14 · Qingtao Pan, Hongzan Sun, Bing Ji, Shuo Li arxiv

Vision Language Model (VLM) has great potential to enhance the quality of pseudo labels in semi-supervised spine segmentation by leveraging textual class prompts to generate segmentation map, but no one has studied it ye…

Enhancing Prompt Injection Attacks to LLMs via Poisoning Alignment

2024-10-18 · Zedian Shao, Hongbin Liu, Jaden Mu, Neil Zhenqiang Gong

In a prompt injection attack, an attacker injects a prompt into the original one, aiming to make an LLM follow the injected prompt to perform an attacker-chosen task. Existing attacks primarily focus on how to blend the …

Self-supervised Fine-tuning for Improved Content Representations by Speaker-invariant Clustering

2023-05-18 · Heng-Jui Chang, Alexander H. Liu, James Glass

Self-supervised speech representation models have succeeded in various tasks, but improving them for content-related problems using unlabeled data is challenging. We propose speaker-invariant clustering (Spin), a novel s…

Acoustic Unit DiscoveryClusteringGPUSelf-Supervised Learning+2

Prompt Injection in Automated Résumé Screening with Large Language Models: Single and Multi-Injection Settings

2026-06-25 · Preet Baxi, Jiannan Xu, Jane Yi Jiang, Stefanus Jasin arxiv

Large language models (LLMs) are increasingly used to screen and rank job applicants, creating incentives for candidates to strategically manipulate algorithmic hiring systems. We study prompt injection in automated résu…