2 papers
cs.CR2026
FragBench: Cross-Session Attacks Hidden in Benign-Looking Fragments
Astha Mehta, Niruthiha Selvanayagam, Cedric Lam +10
An attacker can split a malicious goal into sub-prompts that each look benign on their own and only become harmful in combination. Existing LLM safety benchmarks evaluate prompts o…
physics.soc-ph2022
Is Dogecoin a Viable Investment? Insights from Network and Bubble Effects
Ruoxin Xiao, Xinyu Ying, Hengxu Li +1
We find that three factors: Dogecoin network externalities, momentum, and tweet sentiment that capture the time-series expected Dogecoin returns. Dogecoin returns are exposed to Do…