Llm-Security

New research on LLM jailbreak evaluation, prompt injection taxonomy, federated learning for malware, and coding agents for kernel vulnerabilities.

New papers on MLLM jailbreaks, TCP side-channels, LiDAR spoofing, DoS memory risks, and email aliasing.

New papers on LLM agent robustness, autonomous post-training systems, multi-agent aggregation biases, and delegation governance.

New attacks against aligned LLMs, automatic exploit generation for Node.js, mapping CVEs to MITRE ATT&CK, and cross-privilege BTB side channels.

New papers on bypassing TEE-shielded LLM partition, reverse-engineering iOS location protobufs, localising LLM agent failures, steering LLM agents, and forensic analysis of the HackRF One.