Here's your daily roundup of the most relevant AI and ML news for February 16, 2026. We're also covering 8 research developments. Click through to read the full articles from our curated sources.
Research & Papers
1. Sparse Autoencoders are Capable LLM Jailbreak Mitigators
arXiv:2602.12418v1 Announce Type: cross Abstract: Jailbreak attacks remain a persistent threat to large language model safety. We propose Context-Conditioned Delta Steering (CC-Delta), an SAE-based defense that identifies jailbreak-relevant sparse features by comparing token-level representation...
Source: arXiv - Machine Learning | 18 hours ago
2. Pixel-Based Similarities as an Alternative to Neural Data for Improving Convolutional Neural Network Adversarial Robustness
arXiv:2410.03952v3 Announce Type: replace Abstract: Convolutional Neural Networks (CNNs) excel in many visual tasks but remain susceptible to adversarial attacks-imperceptible perturbations that degrade performance. Prior research reveals that brain-inspired regularizers, derived from neural rec...
Source: arXiv - Machine Learning | 18 hours ago
3. GPU-Fuzz: Finding Memory Errors in Deep Learning Frameworks
arXiv:2602.10478v2 Announce Type: replace-cross Abstract: GPU memory errors are a critical threat to deep learning (DL) frameworks, leading to crashes or even security issues. We introduce GPU-Fuzz, a fuzzer locating these issues efficiently by modeling operator parameters as formal constraints....
Source: arXiv - Machine Learning | 18 hours ago
4. LCSB: Layer-Cyclic Selective Backpropagation for Memory-Efficient On-Device LLM Fine-Tuning
arXiv:2602.13073v1 Announce Type: new Abstract: Memory-efficient backpropagation (MeBP) has enabled first-order fine-tuning of large language models (LLMs) on mobile devices with less than 1GB memory. However, MeBP requires backward computation through all transformer layers at every step, where...
Source: arXiv - Machine Learning | 18 hours ago
5. Self-Supervised Temporal Super-Resolution of Energy Data using Generative Adversarial Transformer
arXiv:2508.10587v4 Announce Type: replace Abstract: To bridge the temporal granularity gap in energy network design and operation based on Energy System Models, resampling of time series is required. While conventional upsampling methods are computationally efficient, they often result in signif...
Source: arXiv - Machine Learning | 18 hours ago
6. TCRL: Temporal-Coupled Adversarial Training for Robust Constrained Reinforcement Learning in Worst-Case Scenarios
arXiv:2602.13040v1 Announce Type: new Abstract: Constrained Reinforcement Learning (CRL) aims to optimize decision-making policies under constraint conditions, making it highly applicable to safety-critical domains such as autonomous driving, robotics, and power grid management. However, existin...
Source: arXiv - Machine Learning | 18 hours ago
7. Sample-Specific Noise Injection For Diffusion-Based Adversarial Purification
arXiv:2506.06027v2 Announce Type: replace-cross Abstract: Diffusion-based purification (DBP) methods aim to remove adversarial noise from the input sample by first injecting Gaussian noise through a forward diffusion process, and then recovering the clean example through a reverse generative pro...
Source: arXiv - Machine Learning | 18 hours ago
8. GT-HarmBench: Benchmarking AI Safety Risks Through the Lens of Game Theory
arXiv:2602.12316v1 Announce Type: new Abstract: Frontier AI systems are increasingly capable and deployed in high-stakes multi-agent environments. However, existing AI safety benchmarks largely evaluate single agents, leaving multi-agent risks such as coordination failure and conflict poorly und...
Source: arXiv - AI | 18 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.