Sparks of artificial general intelligence: Early experiments with gpt-4 S Bubeck, V Chandrasekaran, R Eldan, J Gehrke, E Horvitz, E Kamar, ... arXiv preprint arXiv:2303.12712, 2023 | 2170 | 2023 |

The power of depth for feedforward neural networks R Eldan, O Shamir Conference on learning theory, 907-940, 2016 | 965 | 2016 |

Textbooks are all you need S Gunasekar, Y Zhang, J Aneja, CCT Mendes, A Del Giorno, S Gopi, ... arXiv preprint arXiv:2306.11644, 2023 | 239 | 2023 |

Sparks of artificial general intelligence: Early experiments with GPT-4. arXiv S Bubeck, V Chandrasekaran, R Eldan, J Gehrke, E Horvitz, E Kamar, ... arXiv preprint arXiv:2303.12712, 2023 | 230 | 2023 |

Kernel-based methods for bandit convex optimization S Bubeck, YT Lee, R Eldan Proceedings of the 49th Annual ACM SIGACT Symposium on Theory of Computing …, 2017 | 179* | 2017 |

Textbooks are all you need ii: phi-1.5 technical report Y Li, S Bubeck, R Eldan, A Del Giorno, S Gunasekar, YT Lee arXiv preprint arXiv:2309.05463, 2023 | 162 | 2023 |

Testing for high‐dimensional geometry in random graphs S Bubeck, J Ding, R Eldan, MZ Rácz Random Structures & Algorithms 49 (3), 503-532, 2016 | 151 | 2016 |

Sampling from a log-concave distribution with projected Langevin Monte Carlo S Bubeck, R Eldan, J Lehec Discrete & Computational Geometry 59, 757-783, 2018 | 150 | 2018 |

Thin shell implies spectral gap up to polylog via a stochastic localization scheme R Eldan Geometric and Functional Analysis 23 (2), 532-569, 2013 | 144 | 2013 |

& Zhang, Y.(2023). Sparks of artificial general intelligence: Early experiments with gpt-4 S Bubeck, V Chandrasekaran, R Eldan, J Gehrke, E Horvitz, E Kamar arXiv preprint arXiv:2303.12712, 0 | 102 | |

Tinystories: How small can language models be and still speak coherent english? R Eldan, Y Li arXiv preprint arXiv:2305.07759, 2023 | 99 | 2023 |

Gaussian-width gradient complexity, reverse log-Sobolev inequalities and nonlinear large deviations R Eldan Geometric and Functional Analysis 28 (6), 1548-1596, 2018 | 85 | 2018 |

Multi-scale exploration of convex functions and bandit convex optimization S Bubeck, R Eldan Conference on Learning Theory, 583-589, 2016 | 84 | 2016 |

A two-sided estimate for the Gaussian noise stability deficit R Eldan Inventiones mathematicae 201, 561-624, 2015 | 83 | 2015 |

Approximately gaussian marginals and the hyperplane conjecture R Eldan, B Klartag Concentration, functional inequalities and isoperimetry 545, 55-68, 2011 | 70 | 2011 |

The entropic barrier: a simple and optimal universal self-concordant barrier S Bubeck, R Eldan arXiv preprint arXiv:1412.1587, 2014 | 64 | 2014 |

Localization schemes: A framework for proving mixing bounds for Markov chains Y Chen, R Eldan 2022 IEEE 63rd Annual Symposium on Foundations of Computer Science (FOCS …, 2022 | 63 | 2022 |

Phi-2: The surprising power of small language models M Javaheripi, S Bubeck, M Abdin, J Aneja, S Bubeck, CCT Mendes, ... Microsoft Research Blog, 2023 | 60 | 2023 |

A spectral condition for spectral gap: fast mixing in high-temperature Ising models R Eldan, F Koehler, O Zeitouni Probability theory and related fields 182 (3), 1035-1051, 2022 | 55 | 2022 |

Unveiling transformers with lego: a synthetic reasoning task Y Zhang, A Backurs, S Bubeck, R Eldan, S Gunasekar, T Wagner arXiv preprint arXiv:2206.04301, 2022 | 53 | 2022 |