Quote of the day

A person who never made a mistake never tried anything new.

- Albert Einstein

Contact

Evidence-Grounded LLM Copilot for SOC Alert Triage with Controlled Abstention

Open Cybersecurity

Evidence-Grounded LLM Copilot for SOC Alert Triage with Controlled Abstention

Evidence-Grounded LLM Copilot for SOC Alert Triage with Controlled Abstention

Description

Design an LLM-based security operations copilot that explains and prioritizes alerts only when its conclusions can be traced to source logs, and that abstains when evidence is insufficient.

Details

Context and Problem Statement

SOC analysts face alert overload and high false-positive rates. LLMs can assist with event correlation and reporting, but fluent unsupported explanations can create dangerous overconfidence.

Research Question

How can an LLM be constrained to generate a triage hypothesis only when it is traceable to observed log events, and how should abstention quality be measured?

Proposed Approach

Use RAG over normalized logs, mandatory event-ID citation, MITRE ATT&CK mapping, and a calibrated decision layer that chooses among recommendation, request for additional evidence, and human escalation. Adversarial tests include deceptive logs and prompt injection in attacker-controlled fields.

Expected Contribution

A benchmark jointly measuring triage accuracy, evidence traceability, hallucination, and abstention.

Expected Prototype

A monitoring-stack copilot generating source-backed summaries, confidence levels, reversible action recommendations, and an audit trail.

Datasets

CICIDS2017, CSE-CIC-IDS2018, UNSW-NB15, and controlled testbed logs.

Challenges

Sensitive logs, prompt injection, schema drift, expert evaluation cost, and over-automation risk.

Research Question

How can evidence traceability and calibrated abstention be guaranteed in an LLM-based security alert triage assistant?

Innovation

The project prioritizes reliability over answer coverage by enforcing evidence grounding and treating abstention as a primary performance dimension.

Expected Deliverable

A demonstrable alert-triage copilot, a faithfulness and abstention evaluation protocol, and a failure-mode analysis.

Technologies

Large Language Models Retrieval-Augmented Generation MITRE ATT&CK SIEM and Log Normalization Calibration and Hallucination Detection

Required Skills

  • Information Security and Log Analysis
  • Natural Language Processing and RAG
  • Systems Administration and Containerization
  • Experimental Evaluation Design

Datasets

  • CICIDS2017 / CSE-CIC-IDS2018
  • UNSW-NB15
  • Logs generated in a controlled testbed

Morocco & Africa Relevance

Cybersecurity teams in Morocco and Africa often operate with limited analyst capacity; a reliable copilot can augment small teams while preserving human decision authority.