Increasing inference-time compute enhances the robustness of reasoning models against adversarial attacks, with success rates decreasing as compute increases, suggesting potential for improved adversarial resilience in Large Language Models.

https://arxiv.org/abs//2501.18841

YouTube: https://www.youtube.com/@ArxivPapers

TikTok: https://www.tiktok.com/@arxiv_papers

Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016

Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers