BASIS: Breach-Aware Selective Prompt Injection Shielding with Prefill Attention Probes

arXiv:2608.08027 · cs.CR, cs.LG · Submitted 2026-08-08 · Read on arXiv

Laiqiao Qin, Tianqing Zhu, Longxiang Gao, Wanlei Zhou

cs.CR, cs.LG

Submitted: 2026-08-08

Updated: 2026-08-11

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• AUROC — 52× in this paper · explained in 3 episodes
• Robustness — 42× in this paper · explained in 6 episodes
• Regularization — 18× in this paper · explained in 3 episodes
• AUC — 11× in this paper · explained in 2 episodes
• Prompt Injection Attacks — 9× in this paper · explained in 2 episodes
• Instruction Robustness Score — 7× in this paper
• Role Separation — 6× in this paper · explained in 2 episodes
• Linear probe — 5× in this paper · explained in 3 episodes
• Generalization — 4× in this paper · explained in 14 episodes
• Linear Probes — 4× in this paper · explained in 2 episodes
• Sparsity — 4× in this paper · explained in 5 episodes
• Ablation — 3× in this paper · explained in 1 episode
• Calibration — 3× in this paper · explained in 9 episodes
• Benchmarking — 2× in this paper · explained in 4 episodes
• Consistency — 2× in this paper · explained in 3 episodes
• Large Language Model (LLM) — 2× in this paper · explained in 3 episodes
• Model-Agnostic — 2× in this paper · explained in 2 episodes
• Transformer — 2× in this paper · explained in 2 episodes
• Transformers — 2× in this paper · explained in 2 episodes
• Ablation Studies — 1× in this paper · explained in 1 episode
• Adversarial Robustness — 1× in this paper · explained in 4 episodes
• Clarity — 1× in this paper · explained in 3 episodes
• Compositionality — 1× in this paper · explained in 2 episodes
• Evolutionary Search — 1× in this paper · explained in 3 episodes
• In-Context Learning — 1× in this paper · explained in 7 episodes
• Indirect Prompt Injection — 1× in this paper · explained in 2 episodes
• Instruction Tuning — 1× in this paper · explained in 3 episodes
• Known-Answer Detection — 1× in this paper
• Language Models (LLMs) — 1× in this paper · explained in 2 episodes
• Large Language Models (LLMs) — 1× in this paper · explained in 9 episodes
• Linear Probing — 1× in this paper · explained in 3 episodes
• Papers — 1× in this paper · explained in 3 episodes
• Prompt Engineering — 1× in this paper · explained in 6 episodes
• Structural Constraints — 1× in this paper · explained in 2 episodes

Sources

• A Survey of Large Language Models
• InjecGuard: Benchmarking and Mitigating Over-defense in Prompt Injection Guardrail Models
• Universal and Transferable Adversarial Attacks on Aligned Language Models
• Understanding intermediate layers using linear classifier probes
• In-context Learning and Induction Heads
• Ignore Previous Prompt: Attack Techniques For Language Models
• Llama 2: Open Foundation and Fine-Tuned Chat Models
• Defending Against Indirect Prompt Injection Attacks With Spotlighting
• Qwen3 Technical Report
• The Llama 3 Herd of Models
• Jailbreaking ChatGPT via Prompt Engineering: An Empirical Study

Related papers

• SoK: AI-Augmented Binary Reversing
• Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
• Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
• Estimating the Social Cost of Corporate Data Breaches
• Efficient Fuzzy PSI under One-Sided Assumptions
• Sealing the Audit-Runtime Gap for LLM Skills