System extracts cyber threat attack patterns from full reports accurately
ARBITRA: Robust Document-Level ATT&CK Extraction via Multi-Expert Argumentation
Computation and LanguageCryptography and Security
Summary
Cyber threat reports describe how hackers behave, but turning these reports into clear lists of attack techniques is hard. The paper presents ARBITRA, a system that uses multiple expert AI models working together to suggest possible attack behaviors from entire reports. These suggestions are then checked carefully against the report details using a special reasoning method. This approach gives more accurate and reliable results than previous methods, even when the reports are tricky or confusing.
What this means in practice
- •For cybersecurity teams: Automatically generate structured ATT&CK mappings from full cyber threat reports to improve threat understanding and response.
- •For security software developers: Integrate multi-expert argumentation frameworks to enhance the robustness of automated threat intelligence extraction under adversarial conditions.
Authors
Youngjin Jin, Seungwon Shin
Abstract
Cyber threat intelligence (CTI) reports describe adversary behavior in natural language, but converting them into structured MITRE ATT&CK mappings remains difficult: existing methods decompose reports to sentence-level granularity for extraction, while modern LLMs can work with full reports in a single pass but with noticeable performance degradation. We present ARBITRA, which orchestrates multiple role-specialized LLM experts that propose candidate ATT&CK entities from the full report, which are then verified through a per-candidate quantitative bipolar argumentation framework (QBAF) grounded in document evidence and rationales. We additionally construct reasoning-specialized corpora for fine-tuning so experts can act as effective participants during this verification process. Across ATT&CK extraction benchmarks, ARBITRA reaches a stronger precision-oriented operating point than sentence-level pipelines, recent LLM-based extraction systems, and frontier models including GPT-5.5. Under adversarial perturbations simulating conflicting TTP statements, cross-report contamination, and semantic poisoning, ARBITRA limits mean precision degradation to roughly half that of frontier-model baselines for a more robust extraction.