ranking-agent / ranking-agent.github.io

1 stars 0 forks source link

V2. Ingest sources of curated causal data (e.g. AOPWiki, CAMs) to build a validation set for ARAGORN answers to specific types of questions. #13

Closed mmersmann closed 3 years ago

mmersmann commented 3 years ago

In concert with the evaluation pipeline (see V1), these data will serve to enable rapid development and enhancement of edge-weighting and answer scoring algorithms. These validation examples will also be incorporated into test suites compiled by the Translator consortium for ARS-level testing.