conftrace_

password-locked model

2 papers

Explore in graph

Co-occurring keywords · papers mentioning both

capability elicitation (2) safety evaluation (1) hidden capability (1) fine tuning (1) prompting technique (1) llm alignment (1) large language model (1) dangerous capability (1) fine-tuning evaluation (1) circuit-breaking (1)

Papers

The Elicitation Game: Evaluating Capability Elicitation Techniques ICML 2025 Stress-Testing Capability Elicitation With Password-Locked Models NeurIPS 2024