DeepWeave is a neuro‑symbolic network intrusion detection system that integrates logic rules with neural predictions to provide interpretable, probabilistic evidence for alerts. It automatically generates and validates rules, couples evidence across packet, flow, and hybrid views, and produces proofs that explain rule‑supported detections. In experiments, DeepWeave achieves higher macro‑F1 scores and lower false‑positive rates than a purely neural baseline, even with only 50 labeled packets per attack class, and its explanations are more stable than post‑hoc neural attributions.
By Raoul Saad, Enkeleda Bardhi, Andrea Agiollo, Fernando Kuipers
The paper investigates whether neuro‑symbolic (NeSy) AI models, which combine neural perception with symbolic reasoning, can mitigate backdoor attacks. It presents a systematic evaluation comparing the NeSy framework DeepProbLog to baseline neural networks across eight backdoor settings and four reasoning tasks. Results indicate that NeSy models are generally more robust than pure neural models, but their resilience depends heavily on how strictly the reasoning process is enforced and its alignment with the attack target.
By Marco Antonio Corallo, Andrea Agiollo, Mauro Conti, Alberto Giaretta
arXiv:2610.07334v1 Announce Type: new
Abstract: Interpretability methods for neural networks are predominantly reactive: they analyse activations produced during specific forward passes, requiring kn...
By Krishna Kabra, Constantin Venhoff, Christian Schroeder de Witt
Large language models (LLMs) are vulnerable to backdoor attacks, where hidden triggers induce malicious outputs. Existing defenses generally fall into inference-time detection or training-time mitigation, but face two key limitations.
The paper investigates how dataset imbalance increases vulnerability to backdoor attacks and degrades existing defenses. It introduces Randomized Probability Perturbation (RPP), a black‑box framework that detects poisoned samples using only model output probabilities, providing provable detectability guarantees and a bound on false positives. Experiments on five benchmarks and multiple attacks show RPP outperforms state‑of‑the‑art defenses, especially under class imbalance.
By Miao Lin, Feng Yu, Rui Ning, Lusi Li, Jiawei Chen, Qian Lou, Mengxin Zheng, Chunsheng Xin, Hongyi Wu
arXiv:2507. 05113v3 Announce Type: replace-cross Abstract: Deep Neural Networks (DNNs) are susceptible to backdoor attacks, where adversaries poison training data to implant backdoor into the victim model.
By Binyan Xu, Fan Yang, Xilin Dai, Di Tang, Kehuan Zhang
arXiv:2609.13731v1 Announce Type: new
Abstract: The transition from passive foundation models to autonomous, goal-directed agentic AI systems has introduced unprecedented capabilities by coupling rec...
By Seyedakbar Mostafavi
arXiv:2609.36117v1 Announce Type: new
Abstract: Securing modern AI systems against backdoor attacks remains an open challenge and requires fundamentally principled estimates of the adversary's budget...
By Issam Seddik, Mohamed El Amine Seddik
arXiv:2608. 00566v1 Announce Type: new Abstract: Post-hoc model explainers such as LIME, SHAP, and Integrated Gradients are widely deployed to audit models in high-stakes sensitive domains, including finance, healthcare, and social welfare.
By Niraj Kumar, Harsh Kasyap
MiST (Mid-trained Security Transformer) is a suite of 8B and 32B language models tailored for cybersecurity, achieving strong performance on public benchmarks. The approach uses a mid-training stage that adapts general pre-trained models to the domain by curating a compact, expert-vetted seed corpus and generating high-quality synthetic training data, rather than continual pre-training on large raw text. MiST checkpoints improve mean cybersecurity accuracy by +13.1 and +8.6 absolute percentage points over Qwen baselines for 8B and 32B models, respectively, and provide a stronger initialization for downstream task-specific fine-tuning and reinforcement learning.
By Oded Ovadia, Elad Ben Zaken, Elad Guttman, Orly Moreno Kadosh
arXiv:2607. 19894v1 Announce Type: cross Abstract: Large language models (LLMs) are vulnerable to backdoor attacks, where hidden triggers induce malicious outputs.
By Yuxi Li, Zhibo Zhang, Kailong Wang, Xingshuo Han, Ling Shi, Haoyu Wang
arXiv:2606. 28929v1 Announce Type: cross Abstract: Cybersecurity is a real-life test-bed for many machine learning problems at once, especially when considering modern strides in using Large Language Models (LLMs) to automate processes as ``agents.
By Edward Raff, Maor Ashkenazi, Sagar Samtani, David J. Elkind, Sven Krasser