INFUSE: A Unified Surface Benchmark for Indirect Prompt Injection in Agentic Systems
We introduce INFUSE, an end-to-end benchmark for evaluating IPI throughout the lifecycle of long-horizon agent workflows.
ProofPaper ↗
Key points
- LLM-based agents increasingly execute long-horizon, end-to-end workflows involving shell commands, code modification, dependency installation, and interaction with remote services.
- INFUSE comprises 40 realistic tasks and a structured taxonomy of 56 attack scenarios spanning five threat categories.
- Attacks are injected through realistic interaction surfaces encountered during task execution, extending evaluation beyond attacks placed only in initial prompts or isolated tool interactions.
- We evaluate 19 LLMs from the GPT, Claude, Gemini, Llama, Gemma, and Qwen families, producing 41,824 adversarial trajectories (26,220 without defenses) and 720 benign reference runs.
Sources (1)
- [1]INFUSE: A Unified Surface Benchmark for Indirect Prompt Injection in Agentic SystemsarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 14, 06:26 PM
We introduce INFUSE, an end-to-end benchmark for evaluating IPI throughout the lifecycle of long-horizon agent workflows.
LLM-based agents increasingly execute long-horizon, end-to-end workflows involving shell commands, code modification, dependency installation, and interaction with remote services.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 14, 2026Anuvani-LID: Robust and Calibrated Spoken Language Identification for Indian Languages
- Oct 14, 2026Designing Textual Interfaces To Enhance Llm Reasoning On Audio-Visual Question Answering
- Oct 14, 2026Are You Reading the Right Thing? Rethinking Internal Signals for Prompt-Injection Detection
- Oct 14, 2026Discover AI: A Learning Guide - 51 hands-on labs for learning how AI actually works.
- Oct 13, 2026Run fast, private agents on-device with LFM2.5-24B-A2B
- Oct 13, 2026Gemini 4 Argon: Built to solve your hardest problems
