Here's your daily roundup of the most relevant AI and ML news for April 17, 2026. We're also covering 8 research developments. Click through to read the full articles from our curated sources.
Research & Papers
1. Robustness Analysis of Machine Learning Models for IoT Intrusion Detection Under Data Poisoning Attacks
arXiv:2604.14444v1 Announce Type: cross Abstract: Ensuring the reliability of machine learning-based intrusion detection systems remains a critical challenge in Internet of Things (IoT) environments, particularly as data poisoning attacks increasingly threaten the integrity of model training pip...
Source: arXiv - AI | 10 hours ago
2. Physically-Induced Atmospheric Adversarial Perturbations: Enhancing Transferability and Robustness in Remote Sensing Image Classification
arXiv:2604.14643v1 Announce Type: cross Abstract: Adversarial attacks pose a severe threat to the reliability of deep learning models in remote sensing (RS) image classification. Most existing methods rely on direct pixel-wise perturbations, failing to exploit the inherent atmospheric characteri...
Source: arXiv - Machine Learning | 10 hours ago
3. Benchmarking Linguistic Adaptation in Comparable-Sized LLMs: A Study of Llama-3.1-8B, Mistral-7B-v0.1, and Qwen3-8B on Romanized Nepali
arXiv:2604.14171v1 Announce Type: cross Abstract: Romanized Nepali, the Nepali language written in the Latin alphabet, is the dominant medium for informal digital communication in Nepal, yet it remains critically underresourced in the landscape of Large Language Models (LLMs). This study present...
Source: arXiv - AI | 10 hours ago
4. Route to Rome Attack: Directing LLM Routers to Expensive Models via Adversarial Suffix Optimization
arXiv:2604.15022v1 Announce Type: cross Abstract: Cost-aware routing dynamically dispatches user queries to models of varying capability to balance performance and inference cost. However, the routing strategy introduces a new security concern that adversaries may manipulate the router to consis...
Source: arXiv - Machine Learning | 10 hours ago
5. Best of both worlds: Stochastic & adversarial best-arm identification
arXiv:2604.14860v1 Announce Type: cross Abstract: We study bandit best-arm identification with arbitrary and potentially adversarial rewards. A simple random uniform learner obtains the optimal rate of error in the adversarial scenario. However, this type of strategy is suboptimal when the rewar...
Source: arXiv - Machine Learning | 10 hours ago
6. Improving Clean Accuracy via a Tangent-Space Perspective on Adversarial Training
arXiv:2408.14728v2 Announce Type: replace Abstract: Adversarial training has proven effective in improving the robustness of deep neural networks against adversarial attacks. However, this enhanced robustness often comes at the cost of a substantial drop in accuracy on clean data. In this paper,...
Source: arXiv - Machine Learning | 10 hours ago
7. Prompt Injection as Role Confusion
arXiv:2603.12277v4 Announce Type: replace-cross Abstract: Language models remain vulnerable to prompt injection attacks despite extensive safety training. We trace this failure to role confusion: models infer the source of text based on how it sounds, not where it actually comes from. A command ...
Source: arXiv - AI | 10 hours ago
8. From Memorization to Creativity: LLM as a Designer of Novel Neural Architectures
arXiv:2601.02997v2 Announce Type: replace Abstract: Large language models (LLMs) excel in program synthesis, yet their capacity for neural architecture design -- balancing syntactic reliability, performance, and structural novelty -- remains underexplored. We present a closed-loop architecture s...
Source: arXiv - Machine Learning | 10 hours ago
About This Digest
This digest is automatically curated from leading AI and tech news sources, filtered for relevance to AI security and the ML ecosystem. Stories are scored and ranked based on their relevance to model security, supply chain safety, and the broader AI landscape.
Want to see how your favorite models score on security? Check our model dashboard for trust scores on the top 500 HuggingFace models.