AnaLog

Creators
  • Samuel Ryb
Contributors
Publication date 2022
Description A natural language inference task designed to probe models for analytical and deductive logical reasoning capabilities, controlling for different invalid heuristics the models may adopt instead of learning the desired generalisations.
Publisher GitHub
Organisations
  • Interfacultary Research - Institute for Logic, Language and Computation (ILLC)
Document type Dataset
Related publication AnaLog: Testing Analytical and Deductive Logic Learnability in Language Models
Other links https://github.com/dmg-illc/analog
Permalink to this page
Back