PREVIEW. This is how it will look in an inbox; it was not sent to you. Each real email carries its own unsubscribe link and Keryn's postal address.

Keryn Headlines

Daily, Monday 6 October 2025 · Retrospective · Covers 5 October 2025 00:00 to 6 October 2025 00:00 UTC

Read on the web

Top developments

MulVuln: Enhancing Pre-trained LMs with Shared and Language-Specific Knowledge for Multilingual Vulnerability Detection

arXiv · Research · SAST

Unmasking Backdoors: An Explainable Defense via Gradient-Attention Anomaly Scoring for Pre-trained Language Models

arXiv · Research · Models

Read the Scene, Not the Script: Outcome-Aware Safety for LLMs

arXiv · Research · Models

SAST

QuiLL: An LLM-Based Vulnerability Assessment Framework for the Wild

arXiv · Research · SAST

Models

Rounding-Guided Backdoor Injection in Deep Learning Model Quantization

arXiv · Research · Models

COSMO-RL: Towards Trustworthy LMRMs via Joint Safety and Stability

arXiv · Research · Models

From Poisoned to Aware: Fostering Backdoor Self-Awareness in LLMs

arXiv · Research · Models

AI Security

VortexPIA: Indirect Prompt Injection Attack against LLMs for Efficient Extraction of User Privacy

arXiv · Research · AI Security

AgentTypo: Adaptive Typographic Prompt Injection Attacks against Black-box Multimodal Agents

arXiv · Research · AI Security

Advisories in this window

  • 2 PyPI advisories · 2 moderate
  • 2 npm advisories · 1 moderate · 1 low

All advisories

Titles, links and short excerpts only; every item links to its source. Selection is automatic; the editor's note is AI-drafted and checked as described. Preview: the unsubscribe link and Keryn's postal address are added to each email when it is sent.