TabPFN Always Wins on Benchmarks. Real-World Data Tells a Different Story
The paper says it beats XGBoost without tuning. An experiment on a diabetes dataset with 253K respondents gives us a fresh perspective.
Machine Learning
The paper says it beats XGBoost without tuning. An experiment on a diabetes dataset with 253K respondents gives us a fresh perspective.
Why trees always won on tabular data — and what finally changed that.