SEAL: Reinforcing Global Safety in Mixture-of-Experts through S hared E xpert AL ignment
Open Access
- 14 November 2026
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 48-62
Abstract
No abstract availableFunding Information
- Dutch Research Council (NWO) (024.006.037)
- Dutch Ministry of Education, Culture and Science (OCW) (Sectorplan Bèta en Techniek 2 funding)
This publication has 4 references indexed in Scilit:
- DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learningNature, 2025
- A Closer Look into Mixture-of-Experts in Large Language Models2025
- Not What You've Signed Up For: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection2023
- On the Adversarial Robustness of Mixture of Experts2022