mlsanityData to 7 Oct 2026

Can Neural Networks Understand Logical Entailment?

arXiv 1802.08535 ↗First posted 23 Feb 2018cs.NE cs.AIcited by 24

Not on the map. Why?

Abstract (arXiv)

We introduce a new dataset of logical entailments for the purpose of measuring models' ability to capture and exploit the structure of logical expressions against an entailment prediction task. We use this task to compare a series of architectures which are ubiquitous in the sequence-processing literature, in addition to a new model class---PossibleWorldNets---which computes entailment as a "convolution over possible worlds". Results show that convolutional networks present the wrong inductive bias for this class of problems relative to LSTM RNNs, tree-structured neural networks outperform LSTM RNNs due to their enhanced ability to exploit the syntax of logic, and PossibleWorldNets outperform all benchmarks.

Key facts. Cited by 24 papers in the map (data through 7 Oct 2026). First posted on arXiv 23 Feb 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 9; 2022: 3; 2023: 3; 2024: 3; 2025: 4; 2026: 2.

24 papers.

Sort:

By area:

  1. 2021Proof Artifact Co-training for Theorem Proving with Language Modelscited by 130
  2. 2021Machine Learning for Electronic Design Automation: A Surveycited by 71
  3. 2021LIME: Learning Inductive Bias for Primitives of Mathematical Reasoningcited by 46
  4. 2023Towards LogiGLUE: A Brief Survey and A Benchmark for Analyzing Logical Reasoning Capabilities of Language Modelscited by 35
  5. 2024Logically Consistent Language Models via Neuro-Symbolic Integrationcited by 30
  6. 2021Pushing the Limits of Rule Reasoning in Transformers through Natural Language Satisfiabilitycited by 18
  7. 2023Symbol-LLM: Leverage Language Models for Symbolic System in Visual Human Activity Reasoningcited by 16
  8. 2021Recognizing and Verifying Mathematical Equations using Multiplicative Differential Neural Unitscited by 8
  9. 2022Generalized Quantifiers as a Source of Error in Multilingual NLU Benchmarkscited by 8
  10. 2024Learning Guided Automated Reasoning: A Brief Surveycited by 7
  11. 2023MizAR 60 for Mizar 50cited by 6
  12. 2021Generating Symbolic Reasoning Problems with Transformer GANscited by 5
12 more
  1. 2021Neuro-Symbolic VQA: A review from the perspective of AGI desideratacited by 4
  2. 2021On the Capabilities of Pointer Networks for Deep Deductive Reasoningcited by 4
  3. 2025Giving AI Personalities Leads to More Human-Like Reasoningcited by 3
  4. 2026NL2LOGIC: AST-Guided Translation of Natural Language into First-Order Logic with Large Language Modelscited by 3
  5. 2024BAIT: Benchmarking (Embedding) Architectures for Interactive Theorem-Provingcited by 2
  6. 2022Embeddings as Epistemic States: Limitations on the Use of Pooling Operators for Accumulating Knowledgecited by 1
  7. 2026Zero-Shot Instruction Following in RL via Structured LTL Representationscited by 1
  8. 2021Learning Heuristics for Template-based CEGIS of Loop Invariants with Reinforcement Learningcited by 0
  9. 2022Extending Logical Neural Networks using First-Order Theoriescited by 0
  10. 2025Structured Relational Representationscited by 0
  11. 2025Natural Language Satisfiability: Exploring the Problem Distribution and Evaluating Transformer-based Language Modelscited by 0
  12. 2025Zero-Shot Instruction Following in RL via Structured LTL Representationscited by 0