Copertina

Talk is (Not) Cheap: A Taxonomy and Benchmark Coverage Audit for LLM Attacks

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Iyer, Karthik Raghu, Jamshidi, Yazdan, Bray, Nicholas, Shvets, Alexey A.
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Cryptography and Security Computation and Language
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

The Landscape of Prompt Injection Threats in LLM Agents: From Taxonomy to Analysis
di: Wang, Peiran, et al.
Pubblicazione: (2026)

BiAxisAudit: A Novel Framework to Evaluate LLM Bias Across Prompt Sensitivity and Response-Layer Divergence
di: Gan, Jialing, et al.
Pubblicazione: (2026)

Security Attacks on LLM-based Code Completion Tools
di: Cheng, Wen, et al.
Pubblicazione: (2024)

Tab-MIA: A Benchmark Dataset for Membership Inference Attacks on Tabular Data in LLMs
di: German, Eyal, et al.
Pubblicazione: (2025)

PandaGuard: Systematic Evaluation of LLM Safety against Jailbreaking Attacks
di: Shen, Guobin, et al.
Pubblicazione: (2025)

Counterfactual Evaluation for Blind Attack Detection in LLM-based Evaluation Systems
di: Liu, Lijia, et al.
Pubblicazione: (2025)

Defending LLM Watermarking Against Spoofing Attacks with Contrastive Representation Learning
di: An, Li, et al.
Pubblicazione: (2025)

Beyond Jailbreaking: Auditing Contextual Privacy in LLM Agents
di: Das, Saswat, et al.
Pubblicazione: (2025)

FlexLLM: Exploring LLM Customization for Moving Target Defense on Black-Box LLMs Against Jailbreak Attacks
di: Chen, Bocheng, et al.
Pubblicazione: (2024)

Raccoon: Prompt Extraction Benchmark of LLM-Integrated Applications
di: Wang, Junlin, et al.
Pubblicazione: (2024)

Adversarial Attacks on LLM-as-a-Judge Systems: Insights from Prompt Injections
di: Maloyan, Narek, et al.
Pubblicazione: (2025)

RedTWIZ: Diverse LLM Red Teaming via Adaptive Attack Planning
di: Horal, Artur, et al.
Pubblicazione: (2025)

Self-Disguise Attack: Induce the LLM to disguise itself for AIGT detection evasion
di: Zhou, Yinghan, et al.
Pubblicazione: (2025)

CI-Work: Benchmarking Contextual Integrity in Enterprise LLM Agents
di: Fu, Wenjie, et al.
Pubblicazione: (2026)

Model-Agnostic Lifelong LLM Safety via Externalized Attack-Defense Co-Evolution
di: Zhang, Xiaozhe, et al.
Pubblicazione: (2026)

Personalized Attacks of Social Engineering in Multi-turn Conversations: LLM Agents for Simulation and Detection
di: Kumarage, Tharindu, et al.
Pubblicazione: (2025)

GCG Attack On A Diffusion LLM
di: Neyroud, Ruben, et al.
Pubblicazione: (2025)

LITMUS: Benchmarking Behavioral Jailbreaks of LLM Agents in Real OS Environments
di: Zhang, Chiyu, et al.
Pubblicazione: (2026)

SecurityLingua: Efficient Defense of LLM Jailbreak Attacks via Security-Aware Prompt Compression
di: Li, Yucheng, et al.
Pubblicazione: (2025)

Attestable Audits: Verifiable AI Safety Benchmarks Using Trusted Execution Environments
di: Schnabl, Christoph, et al.
Pubblicazione: (2025)

The Canary's Echo: Auditing Privacy Risks of LLM-Generated Synthetic Text
di: Meeus, Matthieu, et al.
Pubblicazione: (2025)

ContextLeak: Auditing Leakage in Private In-Context Learning Methods
di: Choi, Jacob, et al.
Pubblicazione: (2025)

Auditing Black-Box LLM APIs with a Rank-Based Uniformity Test
di: Zhu, Xiaoyuan, et al.
Pubblicazione: (2025)

Are You Getting What You Pay For? Auditing Model Substitution in LLM APIs
di: Cai, Will, et al.
Pubblicazione: (2025)

Behavioral Canaries: Auditing Private Retrieved Context Usage in RL Fine-Tuning
di: Chen, Chaoran, et al.
Pubblicazione: (2026)

Bag of Tricks: Benchmarking of Jailbreak Attacks on LLMs
di: Xu, Zhao, et al.
Pubblicazione: (2024)

Mapping the Exploitation Surface: A 10,000-Trial Taxonomy of What Makes LLM Agents Exploit Vulnerabilities
di: Mouzouni, Charafeddine
Pubblicazione: (2026)

Domain-Independent Deception: A New Taxonomy and Linguistic Analysis
di: Verma, Rakesh M., et al.
Pubblicazione: (2024)

Probe before You Talk: Towards Black-box Defense against Backdoor Unalignment for Large Language Models
di: Yi, Biao, et al.
Pubblicazione: (2025)

Humanizing the Machine: Proxy Attacks to Mislead LLM Detectors
di: Wang, Tianchun, et al.
Pubblicazione: (2024)

Evading Toxicity Detection with ASCII-art: A Benchmark of Spatial Attacks on Moderation Systems
di: Berezin, Sergey, et al.
Pubblicazione: (2024)

Rubrics as an Attack Surface: Stealthy Preference Drift in LLM Judges
di: Ding, Ruomeng, et al.
Pubblicazione: (2026)

Membership Inference Attacks Against In-Context Learning
di: Wen, Rui, et al.
Pubblicazione: (2024)

Black-Box Guardrail Reverse-engineering Attack
di: Yao, Hongwei, et al.
Pubblicazione: (2025)

LLM-Virus: Evolutionary Jailbreak Attack on Large Language Models
di: Yu, Miao, et al.
Pubblicazione: (2024)

Checkpoint-GCG: Auditing and Attacking Fine-Tuning-Based Prompt Injection Defenses
di: Yang, Xiaoxue, et al.
Pubblicazione: (2025)

Breaking Down the Defenses: A Comparative Survey of Attacks on Large Language Models
di: Chowdhury, Arijit Ghosh, et al.
Pubblicazione: (2024)

Task-Agnostic Detector for Insertion-Based Backdoor Attacks
di: Lyu, Weimin, et al.
Pubblicazione: (2024)

Prompt Stealing Attacks Against Large Language Models
di: Sha, Zeyang, et al.
Pubblicazione: (2024)

AuditGPT: Auditing Smart Contracts with ChatGPT
di: Xia, Shihao, et al.
Pubblicazione: (2024)