Event2Mind: Commonsense Inference on Events, Intents, and Reactions

By Hannah Rashkin, Maarten Sap, Emily Allaway, Noah A. Smith, Yejin Choi. 2018

207 citations recorded by the source corpus.

We investigate a new commonsense inference task: given an event described in a short free-form text (“X drinks coffee in the morning”), a system reasons about the likely intents (“X wants to stay awake”) and reactions (“X feels alert”) of the event’s participants. To support this study, we construct a new crowdsourced corpus of 25,000 event phrases covering a diverse range of everyday events and situations. We report baseline performance on this task, demonstrating that neural encoder-decoder models can successfully compose embedding representations of previously unseen events and reason about the likely intents and reactions of the event participants. In addition, we demonstrate how commonsense inference on people’s intents and reactions can help unveil the implicit gender inequality prevalent in modern movie scripts.

Subfields: Artificial Intelligence

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

A Hybrid Neural Network Model for Commonsense Reasoning (2019, neutral): “, 2016), Event2Mind (Rashkin et al., 2018), SWAG (Zellers et al.”
Social Bias Frames: Reasoning about Social and Power Implications of Language (2019, methodology): “…we collect both categorical as well as free text inferences (described below), inspired by recent efforts in free-text annotations of commonsense knowledge (e.g., Speer and Havasi, 2012; Rashkin et al., 2018; Sap et al., 2019b) and argumentation (Habernal and Gurevych, 2016; Becker et al., 2017).”
“Going on a vacation” takes longer than “Going for a walk”: A Study of Temporal Commonsense Understanding (2019, background): “In terms of “events” commonsense, Rashkin et al. (2018) investigated the intent and reaction of participants of an event, and Zellers et al. (2018) There have also been many works trying to understand time in natural language but not necessarily the commonsense understanding of time.”
Does it Make Sense? And Why? A Pilot Study for Sense Making and Explanation (2019, background): “Event2Mind (Rashkin et al., 2018) proposes a new corpus and task, aiming to ﬁnd out mentioned/unmentioned people’s intents and reactions under various daily circumstances.”
XCOPA: A Multilingual Dataset for Causal Commonsense Reasoning (2020, background): “In recent years, more datasets have been dedicated to other facets of world knowledge (Sakaguchi et al., 2020; Bisk et al., 2020; Bhagavatula et al., 2020; Rashkin et al., 2018; Sap et al., 2019, inter alia ).”
What do Bias Measures Measure? (2021, unknown): “For the task of common sense inference (incorporating common sense knowledge into model inference), Rashkin et al. (2018) analyze the intents of entities involved in an event, ﬁnding gender differences in the intents.”
How Reasonable are Common-Sense Reasoning Tasks: A Case-Study on the Winograd Schema Challenge and SWAG (2018, unknown): “With the release of an increasing number of fine-grained inference tasks aimed at these abilities (Roemmele et al., 2011; Morgenstern et al., 2016; Wang et al., 2018; Rashkin et al., 2018; McCann et al., 2018), the issue of experimental validity in CSR will also become even more important.”
A Knowledge-Enhanced Pretraining Model for Commonsense Story Generation (2020, background): “Incorporating commonsense knowledge is necessary and beneficial for language inference (LoBue and Yates, 2011; Bowman et al., 2015; Rashkin et al., 2018b), reading comprehension (Mihaylov and Frank, 2018; Rashkin et al.”
Analytical Reasoning of Text (2022, background): “108 Commonsense Reasoning There are many re- 109 cent benchmarks that assess the commonsense rea- 110 soning capabilities from different aspects, like so- 111 cial (Rashkin et al., 2018), physics (Talmor et al., 112 2018; Zellers et al., 2019), or temporal (Zhou 113 et al., 2019) aspects.”
On the Evaluation of Common-Sense Reasoning in Natural Language Understanding (2018, background): “eed for proper evaluation protocols. With the release of an increasing number of ﬁne-grained inference tasks aimed at these abilities (Roemmele et al., 2011;Morgenstern et al.,2016; Wang et al.,2018; Rashkin et al., 2018; McCann et al., 2018), the issue of experimental validity in CSR will also beco”
Moral Stories: Situated Reasoning about Norms, Intents, Actions, and their Consequences (2020, methodology): “Our study is, in large parts, motivated by the existing body of research into computational study of social dynamics (Rashkin et al., 2018; Sap et al., 2019a,b, 2020), as well as recent efforts investigating whether NLU / NLG models can reason about norms guiding human behavior.”
Conversational Neuro-Symbolic Commonsense Reasoning (2020, background): “Projects like CYC [7], ConceptNet [8–10] and ATOMIC [11, 12] are examples of such KBs (see [2] for a comprehensive list).”
Commonsense Reasoning for Natural Language Understanding: A Survey of Benchmarks, Resources, and Approaches (2019, methodology): “They may be less restricted to a subset of a large number of category labels, e.g., the Maslow, Reiss, and Plutchik tasks in Story Commonsense (Rashkin et al., 2018a).”
ASER: A Large-scale Eventuality Knowledge Graph (2019, background): “While the aforementioned knowledge bases are annotated by domain experts, ConceptNet4 [25], Event2Mind [39], ProPora [10], and ATOMIC [37] leveraged crowdsourcing platforms or the general public to annotate commonsense knowledge about eventualities, in particular the relations among them.”
Machine Common Sense Concept Paper (2018, background): “Yejin Choi, a new faculty member at UW, has led a series of projects to learn commonsense knowledge from language on the Web (e.g., verb physics [39], event inferences [40], story understanding [41]).”
On Measures of Biases and Harms in NLP (2021, background): “For the task of common sense inference (incorporating common sense knowledge into model inference), Rashkin et al. (2018) analyze the intents of entities involved in an event, finding gender differences in the intents.”
From LSAT: The Progress and Challenges of Complex Reasoning (2021, background): “Recently many benchmarks have been introduced to assess reasoning capabilities over different commonsense knowledge, such as domain-specific knowledge [26, 27, 28], general semantic knowledge [29] and inferential knowledge [30].”