Amelia Glaese

发表论文 7 篇 · 总被引 1308 次 · h-index 7

代表论文

• BrowseComp: A Simple Yet Challenging Benchmark for Browsing Agents (2025 · arXiv.org · 被引 563)
• Measuring short-form factuality in large language models (2024 · arXiv.org · 被引 347)
• Deliberative Alignment: Reasoning Enables Safer Language Models (2024 · Robotics · 被引 294)
• PaperBench: Evaluating AI's Ability to Replicate AI Research (2025 · International Conference on Machine Learning · 被引 256)
• Stress Testing Deliberative Alignment for Anti-Scheming Training (2025 · arXiv.org · 被引 63)
• Trading Inference-Time Compute for Adversarial Robustness (2025 · arXiv.org · 被引 62)