# LLM safety and security

6,765 papers in 12 research directions. Data through 7 Oct 2026.
HTML: https://mlsanity.com/safety/

## Directions (largest first)

- [Safety alignment, refusal and guardrails](https://mlsanity.com/safety/safety-alignment-refusal-and-guardrails/index.md): 1,071 papers, 14 new this week
- [Machine unlearning](https://mlsanity.com/safety/machine-unlearning/index.md): 1,043 papers, 6 new this week
- [Prompt injection and agent security](https://mlsanity.com/safety/prompt-injection-and-agent-security/index.md): 959 papers, 11 new this week
- [Jailbreaking](https://mlsanity.com/safety/jailbreaking/index.md): 911 papers, 5 new this week
- [Watermarking generated content and models](https://mlsanity.com/safety/watermarking-generated-content-and-models/index.md): 717 papers, 6 new this week
- [Backdoor and data poisoning attacks](https://mlsanity.com/safety/backdoor-and-data-poisoning-attacks/index.md): 709 papers, 4 new this week
- [Membership inference attacks](https://mlsanity.com/safety/membership-inference-attacks/index.md): 526 papers, 2 new this week
- [Vision-language and text-to-image safety](https://mlsanity.com/safety/vision-language-and-text-to-image-safety/index.md): 507 papers, 2 new this week
- [Federated unlearning and poisoning](https://mlsanity.com/safety/federated-unlearning-and-poisoning/index.md): 213 papers, 0 new this week
- [Audio jailbreaks and speech backdoors](https://mlsanity.com/safety/audio-jailbreaks-and-speech-backdoors/index.md): 56 papers, 1 new this week
- [Graph unlearning](https://mlsanity.com/safety/graph-unlearning/index.md): 33 papers, 0 new this week
- [Mixture-of-experts routing safety](https://mlsanity.com/safety/mixture-of-experts-routing-safety/index.md): 20 papers, 1 new this week