Showing cs.AIShow all
2 papers · 1 filter
cs.AI2025
Stress Testing Deliberative Alignment for Anti-Scheming Training
Bronson Schoen, Evgenia Nitishinskaya, Mikita Balesni +16
Highly capable AI systems could secretly pursue misaligned goals -- what we call "scheming". Because a scheming AI would deliberately try to hide its misaligned goals and actions,…
cs.AI2025
PaperBench: Evaluating AI's Ability to Replicate AI Research
Giulio Starace, Oliver Jaffe, Dane Sherburn +10
We introduce PaperBench, a benchmark evaluating the ability of AI agents to replicate state-of-the-art AI research. Agents must replicate 20 ICML 2024 Spotlight and Oral papers fro…