mlsanityData to 7 Oct 2026

Robustness of Agentic AI Systems via Adversarially-Aligned Jacobian Regularization

arXiv 2603.04378 ↗First posted 4 Mar 2026cs.LG cs.AI cs.CRcited by 1

On the map: LLM agents and tool use → General LLM agent frameworks and benchmarks

Abstract (arXiv)

As Large Language Models (LLMs) transition into autonomous multi-agent ecosystems, robust minimax training becomes essential yet remains prone to instability when highly non-linear policies induce extreme local curvature in the inner maximization. Standard remedies that enforce global Jacobian bounds are overly conservative, suppressing sensitivity in all directions and inducing a large Price of Robustness. We introduce Adversarially-Aligned Jacobian Regularization (AAJR), a trajectory-aligned approach that controls sensitivity strictly along adversarial ascent directions. We prove that AAJR yields a strictly larger admissible policy class than global constraints under mild conditions, implying a weakly smaller approximation gap and reduced nominal performance degradation. Furthermore, we derive step-size conditions under which AAJR controls effective smoothness along optimization trajectories and ensures inner-loop stability. These results provide a structural theory for agentic robustness that decouples minimax stability from global expressivity restrictions.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use. Builds on 23 of its 28 references that are in the map. First posted on arXiv 4 Mar 2026.

Builds on

23 of 28 references.

Closest ancestors · 3

Same direction, newest first

  1. 2025Small Language Models are the Future of Agentic AI2 Jun
  2. 2023MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework1 Aug
  3. 2023CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Society31 Mar

Other references · 20

  1. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  2. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  3. 2017Improved Training of Wasserstein GANscited by 2,223
  4. 2023AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversationcited by 1,906
  5. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,510
  6. 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,442
  7. 2018Robustness May Be at Odds with Accuracycited by 873
  8. 2024AgentHarm: A Benchmark for Measuring Harmfulness of LLM Agentscited by 406
  9. 2017Certifying Some Distributional Robustness with Principled Adversarial Trainingcited by 333
  10. 2018Regularisation of Neural Networks by Enforcing Lipschitz Continuitycited by 311
  11. 2019On Gradient Descent Ascent for Nonconvex-Concave Minimax Problemscited by 266
  12. 2019Solving a Class of Non-Convex Min-Max Games Using Iterative First Order Methodscited by 148
8 more
  1. 2024Efficient Adversarial Training in LLMs with Continuous Attackscited by 104
  2. 2018Improving DNN Robustness to Adversarial Attacks using Jacobian Regularizationcited by 88
  3. 2019Adversarial Attack Generation Empowered by Min-Max Optimizationcited by 32
  4. 2025LoRA Training Provably Converges to a Low-Rank Global Minimum or It Fails Loudly (But it Probably Won't Fail)cited by 19
  5. 2022Adversarial Machine Learning Attacks Against Video Anomaly Detection Systemscited by 9
  6. 2025LLM-Guided Agentic Object Detection for Open-World Understandingcited by 7
  7. 2026Agentic AI-Empowered Dynamic Survey Frameworkcited by 2
  8. 2026Socially-Weighted Alignment: A Game-Theoretic Framework for Multi-Agent LLM Systemscited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Detecting Adversarial Data via Provable Adversarial Noise Amplificationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Position: The Complexity of Perfect AI Alignment -- Formalizing the RLHF Trilemma
  2. 2024Dissecting Adversarial Robustness of Multimodal LM Agents
  3. 2026AutoRAS: Learning Robust Agentic Systems with Primitive Representations
  4. 2026Sparsity Is Necessary: Polynomial-Time Stability for Agentic LLMs in Large Action Spaces
  5. 2025Stabilizing Off-Policy Training for Long-Horizon LLM Agent via Turn-Level Importance Sampling and Clipping-Triggered Normalization
  6. 2026ARB4WM: An Adversarial Robustness Benchmark for World Models in Continuous Control
  7. 2025TAMAS: Benchmarking Adversarial Risks in Multi-Agent LLM Systems
  8. 2025Online Distributionally Robust LLM Alignment via Regression to Relative Reward