| Training language models to follow instructions with human feedback L Ouyang, J Wu, X Jiang, D Almeida, C Wainwright, P Mishkin, C Zhang, ... Advances in Neural Information Processing Systems 35, 27730-27744, 2022 | 11229 | 2022 |
| GPT-4 technical report OpenAI arXiv, 2023 | 8645* | 2023 |
| Evaluating large language models trained on code M Chen, J Tworek, H Jun, Q Yuan, HPO Pinto, J Kaplan, H Edwards, ... arXiv preprint arXiv:2107.03374, 2021 | 3681 | 2021 |
| Deep reinforcement learning from human preferences PF Christiano, J Leike, T Brown, M Martic, S Legg, D Amodei Advances in Neural Information Processing Systems 30, 4299-4307, 2017 | 3307 | 2017 |
| Let's Verify Step by Step H Lightman, V Kosaraju, Y Burda, H Edwards, B Baker, T Lee, J Leike, ... arXiv preprint arXiv:2305.20050, 2023 | 509 | 2023 |
| Reward learning from human preferences and demonstrations in Atari B Ibarz, J Leike, T Pohlen, G Irving, S Legg, D Amodei Advances in Neural Information Processing Systems, 8011-8023, 2018 | 445 | 2018 |
| AI Safety Gridworlds J Leike, M Martic, V Krakovna, PA Ortega, T Everitt, A Lefrancq, L Orseau, ... arXiv preprint arXiv:1711.09883, 2017 | 358 | 2017 |
| Scalable agent alignment via reward modeling: a research direction J Leike, D Krueger, T Everitt, M Martic, V Maini, S Legg arXiv preprint arXiv:1811.07871, 2018 | 353 | 2018 |
| Recursively summarizing books with human feedback J Wu, L Ouyang, DM Ziegler, N Stiennon, R Lowe, J Leike, P Christiano arXiv preprint arXiv:2109.10862, 2021 | 254 | 2021 |
| Language models can explain neurons in language models S Bills, N Cammarata, D Mossing, H Tillman, L Gao, G Goh, I Sutskever, ... URL https://openaipublic. blob. core. windows. net/neuron-explainer/paper …, 2023 | 220 | 2023 |
| Self-critiquing models for assisting human evaluators W Saunders, C Yeh, J Wu, S Bills, L Ouyang, J Ward, J Leike arXiv preprint arXiv:2206.05802, 2022 | 219 | 2022 |
| Learning to Understand Goal Specifications by Modelling Reward D Bahdanau, F Hill, J Leike, E Hughes, P Kohli, E Grefenstette arXiv preprint arXiv:1806.01946, 2018 | 207* | 2018 |
| Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision C Burns, P Izmailov, JH Kirchner, B Baker, L Gao, L Aschenbrenner, ... arXiv preprint arXiv:2312.09390, 2023 | 193 | 2023 |
| Ranking Templates for Linear Loops J Leike, M Heizmann Logical Methods in Computer Science, 2015 | 104 | 2015 |
| Learning human objectives by evaluating hypothetical behavior S Reddy, A Dragan, S Levine, S Legg, J Leike International Conference on Machine Learning, 8020-8029, 2020 | 89 | 2020 |
| GPT-4o System Card A Hurst, A Lerer, AP Goucher, A Perelman, A Ramesh, A Clark, AJ Ostrow, ... arXiv preprint arXiv:2410.21276, 2024 | 78 | 2024 |
| Institutionalizing ethics in AI through broader impact requirements CEA Prunkl, C Ashurst, M Anderljung, H Webb, J Leike, A Dafoe Nature Machine Intelligence 3 (2), 104-110, 2021 | 76 | 2021 |
| Quantifying Differences in Reward Functions A Gleave, M Dennis, S Legg, S Russell, J Leike arXiv preprint arXiv:2006.13900, 2020 | 71 | 2020 |
| Scaling and evaluating sparse autoencoders L Gao, TD la Tour, H Tillman, G Goh, R Troll, A Radford, I Sutskever, ... arXiv preprint arXiv:2406.04093, 2024 | 69 | 2024 |
| Linear ranking for linear lasso programs M Heizmann, J Hoenicke, J Leike, A Podelski Automated Technology for Verification and Analysis, 365-380, 2013 | 69 | 2013 |
Jeffrey WuAnthropic AI, OpenAIVerified email at anthropic.com
