Skip to content

Latest commit

 

History

History
49 lines (33 loc) · 3.61 KB

README.md

File metadata and controls

49 lines (33 loc) · 3.61 KB
Raw

Awesome Natural Language Reasoning Papers Awesome

A collection of research papers related to Knowledge Reasoning with Natural Language Models.

Mainly contributed and maintained by Zaiqiao, Moy and Songbo.

All PRs and contritutions are very welcome!

Commonsense Reasoning

  1. Bosselut, Antoine, et al. COMET: Commonsense Transformers for Automatic Knowledge Graph Construction. June 2019. paper.
  2. Da, Jeff, et al. Analyzing Commonsense Emergence in Few-Shot Knowledge Models. Sept. 2021. paper.
  3. Liu, Jiacheng, et al. Generated Knowledge Prompting for Commonsense Reasoning. Mar. 2022. paper.
  4. ---. Generated Knowledge Prompting for Commonsense Reasoning. Mar. 2022. paper.
  5. Sap, Maarten, et al. Commonsense Reasoning for Natural Language Processing. 2020. paper.

Symbolic Reasoning

  1. Arabshahi, Forough, et al. Conversational Multi-Hop Reasoning with Neural Commonsense Knowledge and Symbolic Logic Rules. Sept. 2021. paper.
  2. Kassner, Nora, et al. Are Pretrained Language Models Symbolic Reasoners over Knowledge?. 2020. paper.
  3. Minervini, Pasquale, et al. Differentiable Reasoning on Large Knowledge Bases and Natural Language. Apr. 2020. paper.
  4. Yang, Kaiyu, and Jia Deng. Learning Symbolic Rules for Reasoning in Quasi-Natural Language. Nov. 2021. paper.

Arithmetic Reasoning

  1. Rabe, Markus Norman, et al. Mathematical Reasoning via Self-Supervised Skip-Tree Training. 2020. paper.
  2. Wei, Jason, et al. Chain of Thought Prompting Elicits Reasoning in Large Language Models. Jan. 2022. paper.
  3. Welleck, Sean, et al. NaturalProofs: Mathematical Theorem Proving in Natural Language. June 2021. paper.

Survey Papers

Books

  • The Enigma of Reason: A New Theory of Human Understanding

    by Hugo Mercier and Dan Sperber

  • The Book of Why: The New Science of Cause and Effect

    by Judea Pearl and Dana Mackenzie.

Reasoning Datasets

  • ATOMIC 2020: A commonsense knowledge graph with 1.33M everyday inferential knowledge tuples about entities and events.
  • TransOMCS: TransOMCS contains 20 commonsense relations, 101 thousand unique words, and 18.48 million triplets.

Related GitHub Repositories