Cover Image

Are All Prompt Components Value-Neutral? Understanding the Heterogeneous Adversarial Robustness of Dissected Prompt in Large Language Models

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Zheng, Yujia, Li, Tianhao, Huang, Haotian, Zeng, Tianyu, Lu, Jingyu, Chu, Chuangxin, Huang, Yuekai, Jiang, Ziyou, Xiong, Qian, Ge, Yuyao, Li, Mingyang
Format:	Preprint
Published:	2025
Subjects:	Computation and Language Artificial Intelligence Cryptography and Security
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

P‐3.3: Deep Learning in Speech Recognition: A Review of Recent Methods, Applications and Advantages
by: Tianyu Zeng, et al.
Published: (2025)

P‐3.5: A Review on Spatiotemporal Consistency in Artificial Intelligence Text‐to‐Video Generation: Current Status, Challenges and Future Developments
by: Chuangxin Chu, et al.
Published: (2025)

P‐3.4: AI‐Driven Medical Imaging in Cancer Diagnosis: Recent Advances and Trends
by: Tianyu Zeng, et al.
Published: (2025)

Butterfly Effects in Toolchains: A Comprehensive Analysis of Failed Parameter Filling in LLM Tool-Agent Systems
by: Xiong, Qian, et al.
Published: (2025)

Synthetic Poisoning Attacks: The Impact of Poisoned MRI Image on U-Net Brain Tumor Segmentation
by: Li, Tianhao, et al.
Published: (2025)

Know Thy Enemy: Securing LLMs Against Prompt Injection via Diverse Data Synthesis and Instruction-Level Chain-of-Thought Learning
by: Chang, Zhiyuan, et al.
Published: (2026)

Emerging from Ground: Addressing Intent Deviation in Tool-Using Agents via Deriving Real Calls into Virtual Trajectories
by: Xiong, Qian, et al.
Published: (2026)

All Changes May Have Invariant Principles: Improving Ever-Shifting Harmful Meme Detection via Design Concept Reproduction
by: Jiang, Ziyou, et al.
Published: (2026)

SciSafeEval: A Comprehensive Benchmark for Safety Alignment of Large Language Models in Scientific Tasks
by: Li, Tianhao, et al.
Published: (2024)

Prompt-with-Me: in-IDE Structured Prompt Management for LLM-Driven Software Engineering
by: Li, Ziyou, et al.
Published: (2025)

Mimicking the Familiar: Dynamic Command Generation for Information Theft Attacks in LLM Tool-Learning System
by: Jiang, Ziyou, et al.
Published: (2025)

Prompt Exploration with Prompt Regression
by: Feffer, Michael, et al.
Published: (2024)

From Allies to Adversaries: Manipulating LLM Tool-Calling through Adversarial Injection
by: Wang, Haowei, et al.
Published: (2024)

One Shot Dominance: Knowledge Poisoning Attack on Retrieval-Augmented Generation Systems
by: Chang, Zhiyuan, et al.
Published: (2025)

Fall into a Pit, Gain in a Wit: Cognitive-Guided Harmful Meme Detection via Misjudgment Risk Pattern Retrieval
by: Wang, Wenshuo, et al.
Published: (2025)

Understanding the Prompt Sensitivity
by: Liu, Yang, et al.
Published: (2026)

MAPGD: Multi-Agent Prompt Gradient Descent for Collaborative Prompt Optimization
by: Han, Yichen, et al.
Published: (2025)

Do Prompts Really Prompt? Exploring the Prompt Understanding Capability of Whisper
by: Yang, Chih-Kai, et al.
Published: (2024)

PIS: Linking Importance Sampling and Attention Mechanisms for Efficient Prompt Compression
by: Chen, Lizhe, et al.
Published: (2025)

Prism-$Δ$: Differential Subspace Steering for Prompt Highlighting in Large Language Models
by: Ge, Yuyao, et al.
Published: (2026)

Not All Prompts Are Made Equal: Prompt-based Pruning of Text-to-Image Diffusion Models
by: Ganjdanesh, Alireza, et al.
Published: (2024)

Say More with Less: Understanding Prompt Learning Behaviors through Gist Compression
by: Li, Xinze, et al.
Published: (2024)

Heterogeneous Graph Prompt Learning via Adaptive Weight Pruning
by: Wei, Chu-Yuan, et al.
Published: (2025)

One Prompt Fits All: Universal Graph Adaptation for Pretrained Models
by: Huang, Yongqi, et al.
Published: (2025)

Prompt-SAW: Leveraging Relation-Aware Graphs for Textual Prompt Compression
by: Ali, Muhammad Asif, et al.
Published: (2024)

Localization then Neutralization: Gradient-guided Token Suppression against Visual Prompt Injection Attack
by: Zhang, Dongpeng, et al.
Published: (2026)

GAPO: Learning Preferential Prompt through Generative Adversarial Policy Optimization
by: Gu, Zhouhong, et al.
Published: (2025)

PromptCD: Test-Time Behavior Enhancement via Polarity-Prompt Contrastive Decoding
by: Bi, Baolong, et al.
Published: (2026)

Time-Prompt: Integrated Heterogeneous Prompts for Unlocking LLMs in Time Series Forecasting
by: Wang, Zesen, et al.
Published: (2025)

PromptBridge: Cross-Model Prompt Transfer for Large Language Models
by: Wang, Yaxuan, et al.
Published: (2025)

Understanding Catastrophic Interference: On the Identifibility of Latent Representations
by: Li, Yuke, et al.
Published: (2025)

Effective Code Membership Inference for Code Completion Models via Adversarial Prompts
by: Jiang, Yuan, et al.
Published: (2025)

PDZSeg: Adapting the Foundation Model for Dissection Zone Segmentation with Visual Prompts in Robot-assisted Endoscopic Submucosal Dissection
by: Xu, Mengya, et al.
Published: (2024)

Neutral-Reference Prompting for Vision-Language Models
by: Tian, Senmao, et al.
Published: (2026)

Beyond Family Connections: The Sibling Effect in Internal Migration in the 21st Century China
by: Yuyao Li, et al.
Published: (2025)

A Regression Framework for Understanding Prompt Component Impact on LLM Performance
by: Lauziere, Andrew, et al.
Published: (2026)

Harmless Yet Harmful: Neutral Prompting Attacks for Stealthy Hallucination Steering in Agent Skills
by: Hsu, Chia-Yi, et al.
Published: (2026)

Learning to Rewrite Prompts for Personalized Text Generation
by: Li, Cheng, et al.
Published: (2023)

A Unified Framework for Modeling Heterogeneous Financial Data via Dual-Granularity Prompting
by: Lei, Yu, et al.
Published: (2024)

TIP-Editor: An Accurate 3D Editor Following Both Text-Prompts And Image-Prompts
by: Zhuang, Jingyu, et al.
Published: (2024)