Trident : How to Break Deep Reinforcement Learning Cyber Defenses (Agentic)

• URL: http://arxiv.org/abs/2608.04317v1
• Date: Wed, 05 Aug 2026 00:54:57 GMT
• Title: Trident : How to Break Deep Reinforcement Learning Cyber Defenses (Agentic)
• Abstract summary: Trident is an agentic red teaming framework comprising three components.<n>A dataset comprises over 13,000 high-fidelity red-blue interaction trajectories for RLVR.<n>A single trainable 7B planner reduces blue agent defensive performance by an average of 522%.
• Score: 15.20114456320784
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Autonomous cyber defense systems based on Deep Reinforcement Learning (DRL) have attracted significant research attention, yet remain evaluated almost exclusively against static, heuristic red agents, leaving their robustness against adaptive threats critically understudied. Meanwhile, recent advances in Reinforcement Learning with Verifiable Rewards (RLVR) have improved LLM reasoning, but their integration into cybersecurity remains elusive due to the absence of suitable benchmark environments and interaction datasets. To bridge this gap, we introduce Trident, an agentic LLM red teaming framework comprising three components: a dynamic benchmark with isolated sandbox servers spanning CybORG CAGE 4 and CyberWheel, a dataset comprises over 13,000 high-fidelity red-blue interaction trajectories for RLVR, and a ``Code-as-Policy'' RLVR agentic architecture Trident Agentic). The latter reformulates red agent training as a contextual bandit via a tripartite Log Summarizer--Planner--Coder design, where a trainable Planner generates complete attack strategies from compressed execution logs, which a frozen Coder translates into executable Python policies deployed against live DRL defenders. Empirical evaluations reveal a fundamental brittleness in existing defenses: with a single trainable 7B planner, Trident reduces blue agent defensive performance by an average of 522% compared to static red agent baselines while autonomously discovering emergent behaviors such as decoy avoidance and adaptive state prioritization that static heuristics entirely fail to uncover.

Related papers

• Stateful Cooperative Agents Safeguarding LLMs Against Evolving Multi-Turn Attacks [56.45762972211923]
  We present a proactive defense framework for securing LLMs against evolving multi-turn adversarial attacks.<n>It employs a cooperative multi-agent architecture in which specialized agents execute complementary defense strategies.<n>We present the EMRA dataset designed to simulate evolving strategies across multi-turn attacks.
  arXiv  Detail & Related papers  (2026-07-31T14:04:49Z)
• Safety in Self-Evolving LLM Agent Systems: Threats, Amplification, and Case Studies [62.65225104423404]
  Self-evolving agent systems introduce a qualitatively new threat landscape in which adversarial influences become permanently encoded.<n>We present a systematic security and privacy analysis organized around the Module-Lifecycle Attack Surface matrix.<n>Our findings establish that self-evolution converts every known attack category from session-bounded to lineage-persistent.
  arXiv  Detail & Related papers  (2026-06-22T09:23:50Z)
• A Red Teaming Framework for Evaluating Robustness of AI-enabled Security Orchestration, Automation, and Response Systems [8.56054061802862]
  We introduce an autonomous red teaming framework that integrates large language models (LLMs) with reinforcement learning (RL)<n>A hierarchical design combines an LLM-based planner for strategic intent with an RL controller for tactical execution, supported by reward shaping aligned with kill-chain progression.
  arXiv  Detail & Related papers  (2026-05-16T16:46:06Z)
• Dynamic Cyber Ranges [0.3179229570536315]
  We propose Dynamic Cyber Ranges: cyber range environments augmented with LLM-driven Defender agents that harden infrastructure, monitor for intrusions, and respond in real time.<n>Since attacker and defender agents draw from the same underlying model capabilities, Dynamic Cyber Ranges preserve evaluation headroom as models improve.<n> Notably, a smaller, specialized on-premise model (alias2-mini) matched the frontier model's defensive outcomes on multiple scenarios under identical, untuned prompts, and detected the attacker 10x faster on a complex enterprise scenario.
  arXiv  Detail & Related papers  (2026-04-27T08:44:30Z)
• Enhancing Cloud Network Resilience via a Robust LLM-Empowered Multi-Agent Reinforcement Learning Framework [9.460661802313835]
  CyberOps-Bots is a hierarchical multi-agent reinforcement learning framework empowered by Large Language Models (LLMs)<n> Experiments on real cloud datasets show that, compared to state-of-the-art algorithms, CyberOps-Bots maintains network availability 68.5% higher.<n>We will release our framework to the community, facilitating the advancement of robust and autonomous defense in cloud networks.
  arXiv  Detail & Related papers  (2026-01-12T01:25:41Z)
• BlueCodeAgent: A Blue Teaming Agent Enabled by Automated Red Teaming for CodeGen AI [19.047693413887107]
  We propose BlueCodeAgent, an end-to-end blue teaming agent enabled by automated red teaming.<n>Our framework integrates both sides: red teaming generates diverse risky instances, while the blue teaming agent leverages these to detect previously seen and unseen risk scenarios.<n>BlueCodeAgent achieves an average 12.7% F1 score improvement across four datasets in three tasks.
  arXiv  Detail & Related papers  (2025-10-20T22:00:10Z)
• RL Is a Hammer and LLMs Are Nails: A Simple Reinforcement Learning Recipe for Strong Prompt Injection [82.41836544860833]
  We introduce RL-Hammer, a simple recipe for training attacker models that automatically learn to perform strong prompt injections.<n>We propose a set of practical techniques that enable highly effective, universal attacks.<n> RL-Hammer reaches a 98% ASR against GPT-4o and a $72%$ ASR against GPT-5 with the Instruction Hierarchy defense.
  arXiv  Detail & Related papers  (2025-10-06T15:06:04Z)
• AdvEvo-MARL: Shaping Internalized Safety through Adversarial Co-Evolution in Multi-Agent Reinforcement Learning [78.5751183537704]
  AdvEvo-MARL is a co-evolutionary multi-agent reinforcement learning framework that internalizes safety into task agents.<n>Rather than relying on external guards, AdvEvo-MARL jointly optimize attackers and defenders.
  arXiv  Detail & Related papers  (2025-10-02T02:06:30Z)
• Autonomous Cyber Resilience via a Co-Evolutionary Arms Race within a Fortified Digital Twin Sandbox [0.0]
  This paper introduces the Adversarial Resilience Co-evolution (ARC) framework, addressing the "Trinity of Trust"<n>ARC establishes a co-evolutionary arms race within a Fortified Secure Digital Twin.<n>A comprehensive ablation study reveals that the co-evolutionary process itself contributes a 27% performance improvement.
  arXiv  Detail & Related papers  (2025-06-25T03:28:48Z)
• Chasing Moving Targets with Online Self-Play Reinforcement Learning for Safer Language Models [64.47869632167284]
  Conventional language model (LM) safety alignment relies on a reactive, disjoint procedure: attackers exploit a static model, followed by defensive fine-tuning to patch exposed vulnerabilities.<n>This sequential approach creates a mismatch -- attackers overfit to obsolete defenses, while defenders perpetually lag behind emerging threats.<n>We propose Self-RedTeam, an online self-play reinforcement learning algorithm where an attacker and defender agent co-evolve through continuous interaction.
  arXiv  Detail & Related papers  (2025-06-09T06:35:12Z)
• Multi-Objective Reinforcement Learning for Automated Resilient Cyber Defence [0.0]
  Cyber-attacks pose a security threat to military command and control networks, Intelligence, Surveillance, and Reconnaissance (ISR) systems, and civilian critical national infrastructure.<n>The use of artificial intelligence and autonomous agents in these attacks increases the scale, range, and complexity of this threat and the subsequent disruption they cause.<n> Autonomous Cyber Defence (ACD) agents aim to mitigate this threat by responding at machine speed and at the scale required to address the problem.
  arXiv  Detail & Related papers  (2024-11-26T16:51:52Z)
• A Comprehensive Study of the Robustness for LiDAR-based 3D Object Detectors against Adversarial Attacks [84.10546708708554]
  3D object detectors are increasingly crucial for security-critical tasks. It is imperative to understand their robustness against adversarial attacks. This paper presents the first comprehensive evaluation and analysis of the robustness of LiDAR-based 3D detectors under adversarial attacks.
  arXiv  Detail & Related papers  (2022-12-20T13:09:58Z)
• Fixed Points in Cyber Space: Rethinking Optimal Evasion Attacks in the Age of AI-NIDS [70.60975663021952]
  We study blackbox adversarial attacks on network classifiers. We argue that attacker-defender fixed points are themselves general-sum games with complex phase transitions. We show that a continual learning approach is required to study attacker-defender dynamics.
  arXiv  Detail & Related papers  (2021-11-23T23:42:16Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.