@article{improofbenchbenchmarkingaionresearchleve,
  title = {IMProofBench: Benchmarking AI on Research-Level Mathematical Proof Generation},
  author = {Johannes Schmitt and Gergely Bérczi and Jasper Dekoninck and Jeremy Feusi and Tim Gehrunger and Raphael Appenzeller and Pieter Belmans and Alessio Bottini and Jim Bryan and João Camarneiro and Ana Cannas da Silva and Niklas Canova and Ana-Maria Castravet and Timo de Wolff and Claudio Fontanari and Filippo Gaia and Baran Hashemi and Daniel Holmes and David Holmes and Aitor Iribar Lopez and Victor Jaeck and Martina Jørgensen and Steven Kelk and Martijn Kool and Stefan Kuhlmann and Adam Kurpisz and Johannes Lengler and Chiara Meroni and Ingmar Metzler and Martin Möller and Samuel Muñoz-Echániz and David Muñoz-Lahoz and Robert Nowak and Georg Oberdieck and Daniel Platt and Dylan Possamaï and Gabriel Ribeiro and Aluna Rizzoli and Daria Sakhanda and Raúl Sánchez Galán and Zheming Sun and Diaaeldin Taha and Josef Teichmann and Richard P. Thomas and Henk van der Pol and Michel van Garrel and Charles Vial and Ignacio Barros and Benjamin Doerr and Peter Grünwald and Henry Liu and David Martins and Aleksandar Mijatović and Sergej Monavari and Marc Roth and Patrick Schnider and Yannik Schuler and Pim Spelier and Yuuji Tanaka and Ronald van Luijk},
  year = {2025},
  eprint = {2509.26076},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2509.26076},
}