Here's your daily roundup of the most relevant AI and ML news for July 23, 2026. Today's digest includes 1 security-focused story. We're also covering 7 research developments. Click through to read the full articles from our curated sources.
Security & Safety
1. OpenAI Model Hacks into HuggingFace During Cybersecurity Evaluation
Article URL: https://thezvi.substack.com/p/openai-model-hacks-into-huggingface Comments URL: https://news.ycombinator.com/item?id=49020838 Points: 2
Comments: 2
Source: Hacker News - ML Security | 1 hours ago
Research & Papers
2. Towards an Automated Test of LLM Security Knowledge
arXiv:2607.18496v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used for a range of software, hardware and human-centered security tasks. Consequently, LLM performance on security tasks is an active area of measurement and research, often with a focus on identifyi...
Source: arXiv - AI | 10 hours ago
3. Dual Adversarial Fine-tuning for Enhancing Robustness of Large Vision Language Model
arXiv:2607.18958v1 Announce Type: cross Abstract: While Large Vision-Language Models (LVLMs), represented by LLaVA and GPT-4V, have demonstrated remarkable capabilities, their visual inputs remain vulnerable to adversarial attacks, posing significant security risks. Existing defense methods pred...
Source: arXiv - AI | 10 hours ago
4. How Fast Can Reward Models Score? A Systems Study of C++ and PyTorch Inference Runtimes for RLHF
arXiv:2607.19712v1 Announce Type: new Abstract: In RLHF pipelines, reward scoring blocks policy updates. Slow scoring bottlenecks the entire loop, since no update runs until every rollout gets a score. And yet most setups just default to PyTorch eager mode or torch.compile, no one checks if that...
Source: arXiv - Machine Learning | 10 hours ago
5. Adversarial Frontiers: Minimum-Norm Attack Ensembles for Robustness Evaluation
arXiv:2607.19855v1 Announce Type: new Abstract: Adversarial robustness is commonly evaluated with predefined attack ensembles, such as AutoAttack, at a single perturbation budget $\varepsilon$ and on a selective choice of perturbation norms. We argue this formulation is fundamentally limited. Fi...
Source: arXiv - Machine Learning | 10 hours ago
6. Hybrid LSTM-Graph Neural Framework for Robust Financial Fraud Detection and Adversarial Resilience
arXiv:2607.19350v1 Announce Type: cross Abstract: Financial institutions face significant challenges in detecting sophisticated money laundering patterns, such as smurfing and layering, due to extreme data imbalance (0.13% fraud rate) and evolving adversarial evasion tactics. This paper proposes...
Source: arXiv - Machine Learning | 10 hours ago
7. AdvNav: Behavior-Guided Black-Box Adversarial Attacks on Vision-Language Navigation
arXiv:2607.11063v2 Announce Type: replace Abstract: Despite progress in Embodied AI, Vision-and-Language Navigation systems remain vulnerable to adversarial visual disturbances. Most existing methods rely on white-box access to target model gradients, which is often unrealistic for real-world de...
Source: arXiv - AI | 10 hours ago
8. Reward-Aware Population Scaling of Evolutionary Strategies in LLM Fine-Tuning
arXiv:2607.19408v1 Announce Type: new Abstract: Using Evolutionary Strategies (ES) for fine-tuning large language models is attractive because it is memory-efficient, parallel, and compatible with black-box or discrete rewards. Yet its population-size conclusions conflict sharply: fine-tuning wi...
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.