Can Language Models perform Abductive Commonsense Reasoning?
On the map: LLM reasoning → Reasoning benchmarks and evaluation
Abstract (arXiv)
Abductive Reasoning is a task of inferring the most plausible hypothesis given a set of observations. In literature, the community has approached to solve this challenge by classifying/generating a likely hypothesis that does not contradict with a past observation and future observation. Some of the most well-known benchmarks that tackle this problem are aNLI and aNLG (pronounced as alpha-NLI and alpha-NLG). In this report, I review over some of the methodologies that were attempted to solve this challenge, re-implement the baseline models, and analyze some of the weaknesses that current approaches have. The code and the re-implemented results are available at this link.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning. Builds on 20 of its 27 references that are in the map. First posted on arXiv 7 Jul 2022.
Builds on
20 of 27 references.
References · 20
- 2020Language Models are Few-Shot Learnerscited by 32,199
- 2019HellaSwag: Can a Machine Really Finish Your Sentence?cited by 4,018
- 2019PIQA: Reasoning about Physical Commonsense in Natural Languagecited by 2,818
- 2020Shortcut Learning in Deep Neural Networkscited by 1,928
- 2018CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledgecited by 1,793
- 2019Plug and Play Language Models: A Simple Approach to Controlled Text Generationcited by 793
- 2019CTRL: A Conditional Transformer Language Model for Controllable Generationcited by 754
- 2019COMET: Commonsense Transformers for Automatic Knowledge Graph Constructioncited by 450
- 2019Explain Yourself! Leveraging Language Models for Commonsense Reasoningcited by 367
- 2019Abductive Commonsense Reasoningcited by 300
- 2022COLD Decoding: Energy-based Constrained Text Generation with Langevin Dynamicscited by 159
- 2019"Going on a vacation" takes longer than "Going for a walk": A Study of Temporal Commonsense Understandingcited by 121
8 more
- 2019Counterfactual Story Reasoning and Generationcited by 87
- 2021Understanding by Understanding Not: Modeling Negation in Language Modelscited by 73
- 2020Back to the Future: Unsupervised Backprop-based Decoding for Counterfactual and Abductive Commonsense Reasoningcited by 65
- 2020Like hiking? You probably enjoy nature: Persona-grounded Dialog with Commonsense Expansionscited by 49
- 2020Connecting the Dots: A Knowledgeable Path Generator for Commonsense Question Answeringcited by 44
- 2020WinoWhy: A Deep Diagnosis of Essential Commonsense Knowledge for Answering Winograd Schema Challengecited by 43
- 2019Attention Is (not) All You Need for Commonsense Reasoningcited by 13
- 2021Progressive Open-Domain Response Generation with Multiple Controllable Attributescited by 6
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2023Abductive Commonsense Reasoning Exploiting Mutually Exclusive Explanations
- 2026Abductive Reasoning with Probabilistic Commonsense
- 2026Wiring the 'Why': A Unified Taxonomy and Survey of Abductive Reasoning in LLMs
- 2025GEAR: A General Evaluation Framework for Abductive Reasoning
- 2025What Kind of Reasoning (if any) is an LLM actually doing? On the Stochastic Nature and Abductive Appearance of Large Language Models
- 2023UNcommonsense Reasoning: Abductive Reasoning about Uncommon Situations
- 2025Abductive Inference in Retrieval-Augmented Language Models: Generating and Validating Missing Premises
- 2022AbductionRules: Training Transformers to Explain Unexpected Inputs