# ICON: Indirect Prompt Injection Defense for Agents based on Inference-Time Correction

arXiv: https://arxiv.org/abs/2602.20708 · HTML: https://mlsanity.com/paper/2602.20708/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Placed in Agent threats, guardrails and governance, one of 16 directions in LLM agents and tool use.
- Builds on 25 of its 37 references that are in the map.
- First posted on arXiv 24 Feb 2026.

## Placed in

- LLM agents and tool use / [Agent threats, guardrails and governance](https://mlsanity.com/agents/agent-threats-guardrails-and-governance/)
- LLM safety and security / [Prompt injection and agent security](https://mlsanity.com/safety/prompt-injection-and-agent-security/)

## Builds on (most cited first, 25 of 25)

- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks](https://mlsanity.com/paper/2312.14238/) (2023)
- [Not what you've signed up for: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection](https://mlsanity.com/paper/2302.12173/) (2023)
- [Refusal in Language Models Is Mediated by a Single Direction](https://mlsanity.com/paper/2406.11717/) (2024)
- [Ignore Previous Prompt: Attack Techniques For Language Models](https://mlsanity.com/paper/2211.09527/) (2022)
- [Baseline Defenses for Adversarial Attacks Against Aligned Language Models](https://mlsanity.com/paper/2309.00614/) (2023)
- [Function Vectors in Large Language Models](https://mlsanity.com/paper/2310.15213/) (2023)
- [SecAlign: Defending Against Prompt Injection with Preference Optimization](https://mlsanity.com/paper/2410.05451/) (2024)
- [MiniCPM-V 4.5: Cooking Efficient MLLMs via Architecture, Data, and Training Recipe](https://mlsanity.com/paper/2509.18154/) (2025)
- [Qwen3Guard Technical Report](https://mlsanity.com/paper/2510.14276/) (2025)
- [Lookback Lens: Detecting and Mitigating Contextual Hallucinations in Large Language Models Using Only Attention Maps](https://mlsanity.com/paper/2407.07071/) (2024)
- [Adaptive Attacks Break Defenses Against Indirect Prompt Injection Attacks on LLM Agents](https://mlsanity.com/paper/2503.00061/) (2025)
- [Dissecting Adversarial Robustness of Multimodal LM Agents](https://mlsanity.com/paper/2406.12814/) (2024)
- [Attention Tracker: Detecting Prompt Injection Attacks in LLMs](https://mlsanity.com/paper/2411.00348/) (2024)
- [Tell Your Model Where to Attend: Post-hoc Attention Steering for LLMs](https://mlsanity.com/paper/2311.02262/) (2023)
- [Successor Heads: Recurring, Interpretable Attention Heads In The Wild](https://mlsanity.com/paper/2312.09230/) (2023)
- [MELON: Provable Defense Against Indirect Prompt Injection Attacks in AI Agents](https://mlsanity.com/paper/2502.05174/) (2025)
- [Word Embeddings Are Steers for Language Models](https://mlsanity.com/paper/2305.12798/) (2023)
- [Controllable Text Generation for Large Language Models: A Survey](https://mlsanity.com/paper/2408.12599/) (2024)
- [AnyAttack: Towards Large-scale Self-supervised Adversarial Attacks on Vision-language Models](https://mlsanity.com/paper/2410.05346/) (2024)
- [Goal-guided Generative Prompt Injection Attack on Large Language Models](https://mlsanity.com/paper/2404.07234/) (2024)
- [Indirect Prompt Injections: Are Firewalls All You Need, or Stronger Benchmarks?](https://mlsanity.com/paper/2510.05244/) (2025)
- [Automating Steering for Safe Multimodal Large Language Models](https://mlsanity.com/paper/2507.13255/) (2025)

## Built on by (most cited first, 10 of 10)

- [Toward Secure LLM Agents: Threat Surfaces, Attacks, Defenses, and Evaluation](https://mlsanity.com/paper/2606.10749/) (2026)
- [Constraining Host-Level Abuse in Self-Hosted Computer-Use Agents via TEE-Backed Isolation](https://mlsanity.com/paper/2605.06393/) (2026)
- [Readable Before Actionable: Causal Tracing of Indirect Prompt Injection](https://mlsanity.com/paper/2610.05295/) (2026)
- [Depth-Dependent Indirect Prompt Injection in Tool-Calling ReAct Agents: Injection Depth, Payload Framing, and Turn-Budget Sensitivity](https://mlsanity.com/paper/2605.30686/) (2026)
- [SynChain: Inducing Computer-Use Agent Systems to Construct Their Own Attack Chains](https://mlsanity.com/paper/2608.06862/) (2026)
- [ActGuard: Pre-execution Action Auditing against Indirect Prompt Injection in LLM Agents](https://mlsanity.com/paper/2609.14987/) (2026)
- [When the Agent Becomes the Kernel: A Systematization of Security on the Path to AI-Native Operating Systems](https://mlsanity.com/paper/2609.23700/) (2026)
- [CounterSteer: Suppressing Indirect Prompt Injection with Activation Steering](https://mlsanity.com/paper/2609.36570/) (2026)
- [ToolFence: Fine-Grained Authorization for Secure Tool-Using LLM Agents](https://mlsanity.com/paper/2609.37196/) (2026)
- [TRACER: Trace-Regularized Training Against Indirect Prompt Injection](https://mlsanity.com/paper/2610.19133/) (2026)

## Abstract (arXiv)

Large Language Model (LLM) agents are susceptible to Indirect Prompt Injection (IPI) attacks, where malicious instructions in retrieved content hijack the agent's execution. Existing defenses typically rely on strict filtering or refusal mechanisms, which suffer from a critical limitation: over-refusal, prematurely terminating valid agentic workflows. We propose ICON, a probing-to-mitigation framework that neutralizes attacks while preserving task continuity. Our key insight is that IPI attacks leave distinct over-focusing signatures in the latent space. We introduce a Latent Space Trace Prober to detect attacks based on high intensity scores. Subsequently, a Mitigating Rectifier performs surgical attention steering that selectively manipulate adversarial query key dependencies while amplifying task relevant elements to restore the LLM's functional trajectory. Extensive evaluations on multiple backbones show that ICON achieves a competitive 0.4% ASR, matching commercial grade detectors, while yielding a over 50% task utility gain. Furthermore, ICON demonstrates robust Out of Distribution(OOD) generalization and extends effectively to multi-modal agents, establishing a superior balance between security and efficiency.