Showing cs.CLShow all
2 papers · 1 filter
cs.CL2024
Can Models Help Us Create Better Models? Evaluating LLMs as Data Scientists
MichaÅ Pietruszka, Åukasz Borchmann, Aleksander JÄdrosz +1
We present a benchmark for large language models designed to tackle one of the most knowledge-intensive tasks in data science: writing feature engineering code, which requires doma…
cs.CL2024
Arctic-TILT. Business Document Understanding at Sub-Billion Scale
Åukasz Borchmann, MichaÅ Pietruszka, Wojciech JaÅkowski +13
The vast portion of workloads employing LLMs involves answering questions grounded on PDF or scan content. We introduce the Arctic-TILT achieving accuracy on par with models 1000$\…