TabPFN and TabICL vs. tuned XGBoost: the model that doesn't train won 14/14
Thread
Unofficial Hacker News client; not affiliated with Y Combinator.
TabPFN and TabICL vs. tuned XGBoost: the model that doesn't train won 14/14
Unofficial Hacker News client; not affiliated with Y Combinator.
rdedev · · focus · HN ↗
In drug property prediction domain, tabular foundation models coupled with another foundation model for molecules are pretty close to being the state of art.
Btw the article makes heavy use of AI or is written in that way A lot of unnecessary dramatic flair that gets very tiring
NordStreamYacht · · focus · HN ↗
Why "Every number is a measured one" and not "every parameter is measured?"
Stopped reading at that point.
LLMs are like autotune. Imagine the Rolling Stones auto-tuned.
3abiton · · focus · HN ↗