Challenge: AGENTVIGIL is a black-box optimization framework to exploit indirect prompt injection vulnerabilities . indirect prompts compromise the core of LLM agents by manipulating contextual information rather than direct user prompts.
Approach: They propose a black-box optimization framework to exploit indirect prompt injection vulnerabilities . they use a Monte Carlo tree-based algorithm to iteratively refine inputs .
Outcome: The proposed framework achieves 71% and 70% success rates against two public benchmarks .

Similar Papers

The Dangers of Indirect Prompt Injection Attacks on LLM-based Autonomous Web Navigation Agents: A Demonstration (2025.emnlp-demos)

Copied to clipboard

Challenge: Large Language Model (LLM)-integrated applications are becoming more popular to support, augment, and automate tasks.
Approach: They propose to embed universal adversarial triggers in webpage HTML to hijack agents . they also use a browser-gym agent powered by Llama-3.1 to test their system .
Outcome: The proposed system software is released under the MIT License .
SIRAJ: Diverse and Efficient Red-Teaming for LLM Agents via Distilled Structured Reasoning (2026.findings-eacl)

Copied to clipboard

Challenge: Existing red-teaming frameworks do not cover all the risks associated with arbitrary black-box LLMs.
Approach: They propose a generic red-teaming framework for arbitrary black-box LLM agents that iteratively constructs and refines model-based adversarial attacks based on the execution trajectories of former attempts.
Outcome: The proposed model improves attack success rate by 100%, surpassing the 671B Deepseek-R1 model.
VIGIL: Defending LLM Agents Against Tool-Stream Injection via Verify-Before-Commit (2026.acl-long)

Copied to clipboard

Challenge: Existing defenses for indirect prompt injection are limited by static protection mechanisms . existing models prioritize injected rules due to strict alignment, whereas static protections sever the feedback loop required for adaptive reasoning.
Approach: They propose a framework that shifts the paradigm from restrictive isolation to a verify-before-commit protocol.
Outcome: The proposed framework outperforms state-of-the-art dynamic defenses by reducing the attack success rate by over 22% while more thandoubling utility under attack compared to static baselines.
InjecAgent: Benchmarking Indirect Prompt Injections in Tool-Integrated Large Language Model Agents (2024.findings-acl)

Copied to clipboard

Challenge: Recent work has embodied LLMs as agents, but external content introduces the risk of indirect prompt injection (IPI) attacks.
Approach: They propose a benchmark to assess the vulnerability of tool-integrated LLM agents to IPI attacks.
Outcome: The proposed benchmark assesses the vulnerability of tool-integrated LLM agents to IPI attacks.
Agent vs. Agent: Automated Data Generation and Red-Teaming for Custom Agentic Workflows (2025.emnlp-industry)

Copied to clipboard

Challenge: Existing red-teaming frameworks like AgentHarm use static prompts and hardcoded toolsets .
Approach: They propose a red-teaming framework that generates adversarial tasks and evaluation functions tailored to arbitrary toolsets and uses iterative prompt refinement with self-reflection to develop more effective attacks.
Outcome: The proposed approach achieves 162% increase in attack success rate on o4-mini and 86% success on gemini 2.5 Pro.
Attack Prompt Generation for Red Teaming and Defending Large Language Models (2023.findings-emnlp)

Copied to clipboard

Challenge: Existing studies construct attack prompts via manual or automatic methods, but these methods have limitations on cost and quality.
Approach: They propose an attack framework to instruct LLMs to mimic human-generated prompts through in-context learning and a defense framework that fine-tunes victim LLM's through iterative interactions with the attack framework.
Outcome: The proposed approach is based on experiments on different LLMs to evaluate their effectiveness against red teaming attacks.
The Task Shield: Enforcing Task Alignment to Defend Against Indirect Prompt Injection in LLM Agents (2025.acl-long)

Copied to clipboard

Challenge: Large Language Model (LLM) agents are becoming conversational assistants . indirect prompt injection attacks pose a critical threat to these systems .
Approach: They propose a novel and orthogonal perspective that reframes agent security . they propose 'task shield' that verifies whether each instruction and tool call contributes to user objectives .
Outcome: The proposed defense reduces attack success rates while maintaining high task utility on the AgentDojo benchmark.
Automated Progressive Red Teaming (2025.coling-main)

Copied to clipboard

Challenge: Automated red teaming (ART) is effective but time-consuming, costly and lacks scalability.
Approach: They propose an automated red teaming framework that generates adversarial prompts to expose LLM vulnerabilities.
Outcome: The proposed framework explores and exploits LLM vulnerabilities through multi-round interactions.
MulVul: Retrieval-augmented Multi-Agent Code Vulnerability Detection via Cross-Model Prompt Evolution (2026.acl-long)

Copied to clipboard

Challenge: Large Language Models (LLMs) struggle to automate real-world vulnerability detection due to the heterogeneity of vulnerability patterns and manual prompt engineering for massive weakness categories is unscalable.
Approach: They propose a retrieval-augmented multi-agent framework for precise and broad-coverage vulnerability detection using a coarse-to-fine strategy.
Outcome: The proposed framework outperforms the baseline model on 130 CWE types and achieves 34.79% Macro-F1 performance.
When Prompt Optimization Becomes Jailbreaking: Adaptive Red-Teaming of Large Language Models (2026.eacl-srw)

Copied to clipboard

Challenge: Existing safety evaluations rely on fixed collections of harmful prompts . such attacks span single-shot prompts, multi-turn interactions, cross-lingual settings .
Approach: They propose to use black-box prompt optimization techniques to search for safety failures . they use GPT-5.1 to optimize for a continuous danger score .
Outcome: The proposed approach reduces effective safeguards for large language models . the average danger score of Qwen 3 8B increases from 0.09 in its baseline setting to 0.79 after optimization.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations