Complex Reasoning over Logical Queries on Commonsense Knowledge Graphs

Event commonsense reasoning requires the ability to reason about the relationship between events, as well as infer implicit context underlying that relationship. However, data scarcity makes it challenging for language models to learn to generate commonsense inferences for contexts and questions involving interactions between complex events. To address this demand, we present COM 2 (COMplex COMmonsense), a new dataset created by sampling multi-hop logical queries (e.g., the joint effect or cause of both event A and B, or the effect of the effect of event C) from an existing commonsense knowledge graph (CSKG), and verbalizing them using handcrafted rules and large language models into multiple-choice and text generation questions. Our experiments show that language models trained on COM 2 exhibit significant improvements in complex reasoning ability, resulting in enhanced zero-shot performance in both indomain and out-of-domain tasks for question answering and generative commonsense reasoning, without expensive human annotations. 1 * Work done during internship at EPFL. 1 Code and data are available at https://github.com/ tqfang/complex-commonsense-reasoning find new things to do PersonX goes skydiving PersonX gets tired of it (the intention of PersonX) xIntent xWant (then PersonX wants to) Verbalization After getting tired of it, PersonX goes skydiving PersonX is living a boring life. 🤖 LLM-added context Rule-based discourse Question: What's both the intention of PersonX going skydiving and what X wants to do after PersonX getting tired of it? Answer: find new things to do 𝑞 𝑉 ? = 𝑉 ? : xIntent X goes sky diving , 𝑉 ? ∧ xWant (X gets tired of it, 𝑉 ?