Datasets:

ulamai
/

STEM-RL-Tasks

Modalities:

Formats:

Languages:

Size:

Tags:

scientific-reasoning

reinforcement-learning

verifiable-rewards

numerical-reasoning

Libraries:

Dataset card Data Studio Files Files and versions

STEM-RL-Tasks / tasks /ENG3 /solution /solution.json

przemekch's picture

Publish STEM RL Tasks: 40 tasks across eight domains with verifiers and recorded attempts

6395f35 verified 3 days ago

Raw History Blame Contribute Delete

61 Bytes

{
  "ground_state_mechanical_kinetic_energy_mev": 49.56169
}