Recent (6 mo)
1With code
0Influential cites
0Benchmarked
0Publications per year
126
Top categories
AI×1Crypto×1Software Eng.×1
Frequent co-authors
Research Timeline
2026
Do Agents Dream of Root Shells? Partial-Credit Evaluation of LLM Agents in Capture the Flag Challenges
The paper introduces DeepRed, a new benchmark for evaluating LLM agents in realistic CTF challenges, finding that current agents are limited, achieving only 35% average checkpoint completion.
Highlighted terms show continued research focus across papers