Hacker News new | ask | show | jobs
Selection-Inference: Exploiting Language Models Interpretable Logical Reasoning (arxiv.org)
4 points by silencedogood3 1493 days ago
2 comments
trenchgun 1492 days ago
Twitter thread by one of authors: https://twitter.com/mpshanahan/status/1527686538720595969
link
trenchgun 1492 days ago
So, they got neural networks to do multistep logical reasoning by adding the result of single step reasoning to the knowledge base?

What is the performance compared to symbolic reasoning?

link