English

Towards General Natural Language Understanding with Probabilistic Worldbuilding

Computation and Language 2021-12-22 v2

Abstract

We introduce the Probabilistic Worldbuilding Model (PWM), a new fully-symbolic Bayesian model of semantic parsing and reasoning, as a first step in a research program toward more domain- and task-general NLU and AI. Humans create internal mental models of their observations which greatly aid in their ability to understand and reason about a large variety of problems. In PWM, the meanings of sentences, acquired facts about the world, and intermediate steps in reasoning are all expressed in a human-readable formal language, with the design goal of interpretability. PWM is Bayesian, designed specifically to be able to generalize to new domains and new tasks. We derive and implement an inference algorithm that reads sentences by parsing and abducing updates to its latent world model that capture the semantics of those sentences, and evaluate it on two out-of-domain question-answering datasets: (1) ProofWriter and (2) a new dataset we call FictionalGeoQA, designed to be more representative of real language but still simple enough to focus on evaluating reasoning ability, while being robust against heuristics. Our method outperforms baselines on both, thereby demonstrating its value as a proof-of-concept.

Keywords

Cite

@article{arxiv.2105.02486,
  title  = {Towards General Natural Language Understanding with Probabilistic Worldbuilding},
  author = {Abulhair Saparov and Tom M. Mitchell},
  journal= {arXiv preprint arXiv:2105.02486},
  year   = {2021}
}

Comments

Accepted to TACL; pre-MIT Press publication version

R2 v1 2026-06-24T01:49:45.074Z