27 citations · 60 across the 8 of their papers we have counts for
8 papers
MirrorCode: AI can rebuild entire programs from behavior alone
Tom Adamczewski, David Owen, David Rein +4
AI models are rapidly improving at autonomous coding, as shown by benchmark progress and one-off demonstrations such as AI implementing a C compiler. However, existing coding bench…
Separate the Wheat from the Chaff: Winnowing Down Divergent Views in Retrieval Augmented Generation
Song Wang, Zihan Chen, Peng Wang +5
Retrieval-augmented generation (RAG) enhances large language models (LLMs) by integrating external knowledge sources to address their limitations in accessing up-to-date or special…
Measuring the Impact of Early-2025 AI on Experienced Open-Source Developer Productivity
Joel Becker, Nate Rush, Elizabeth Barnes +1
Despite widespread adoption, the impact of AI tools on software development in the wild remains understudied. We conduct a randomized controlled trial (RCT) to understand how AI to…
HCAST: Human-Calibrated Autonomy Software Tasks
David Rein, Joel Becker, Amy Deng +19
To understand and predict the societal impacts of highly autonomous AI systems, we need benchmarks with grounding, i.e., metrics that directly connect AI performance to real-world…
Measuring AI Ability to Complete Long Software Tasks
Thomas Kwa, Ben West, Joel Becker +23
Despite rapid progress on AI benchmarks, the real-world meaning of benchmark performance remains unclear. To quantify the capabilities of AI systems in terms of human capabilities,…
Training Language Models to Win Debates with Self-Play Improves Judge Accuracy
Samuel Arnesen, David Rein, Julian Michael
We test the robustness of debate as a method of scalable oversight by training models to debate with data generated via self-play. In a long-context reading comprehension task, we…
GPQA: A Graduate-Level Google-Proof Q&A Benchmark
David Rein, Betty Li Hou, Asa Cooper Stickland +5
We present GPQA, a challenging dataset of 448 multiple-choice questions written by domain experts in biology, physics, and chemistry. We ensure that the questions are high-quality…
Debate Helps Supervise Unreliable Experts
Julian Michael, Salsabila Mahdi, David Rein +4
As AI systems are used to answer more difficult questions and potentially help create new knowledge, judging the truthfulness of their outputs becomes more difficult and more impor…