// the find
benedekrozemberczki/awesome-gradient-boosting-papers
A curated list of gradient boosting research papers with implementations.
A bibliography of gradient boosting research papers from major ML/AI conferences (NeurIPS, ICML, KDD, etc.) spanning 2014–2025, with links to papers and implementations where they exist. It's for researchers or practitioners who want to find the original papers behind techniques like XGBoost, LightGBM, CatBoost, or their more obscure cousins — fairness-constrained boosting, federated boosting, probabilistic boosting, and so on. Not a tutorial, not a library.
Conference coverage is genuinely broad — not just the ML venues but also CV (CVPR, ICCV), NLP (ACL, EMNLP), and data mining (KDD, CIKM, SIGIR), which means you'll find boosting papers you wouldn't find on a pure ML reading list. Updated through early 2025 so recent work like NRGBoost and GBRL is included. Implementation links are included where they exist, which is useful since many of these papers have no obvious official repo. The maintainer is the paper author on several entries (BoostedFactorization, etc.), so the list isn't purely mechanical.
A substantial fraction of entries have no code link — the list doesn't distinguish between 'no code exists' and 'code exists but wasn't found', which matters if you're trying to reproduce something. There's no tagging or categorization beyond year and conference, so finding all papers on federated boosting or fairness means reading linearly through the whole thing. The 'awesome.py' file hints at automation but the repo itself is just a README — no search, no filtering, no citation counts. Some older papers (2014–2016) link to ACM or IEEE pages that are paywalled without affiliation access.