Dense Event Ordering with a Multi-Pass Architecture

By Nathanael Chambers, Taylor Cassidy, Bill McDowell, Steven Bethard. 2014

197 citations recorded by the source corpus.

The past 10 years of event ordering research has focused on learning partial orderings over document events and time expressions. The most popular corpus, the TimeBank, contains a small subset of the possible ordering graph. Many evaluations follow suit by only testing certain pairs of events (e.g., only main verbs of neighboring sentences). This has led most research to focus on specific learners for partial labelings. This paper attempts to nudge the discussion from identifying some relations to all relations. We present new experiments on strongly connected event graphs that contain ∼10 times more relations per document than the TimeBank. We also describe a shift away from the single learner to a sieve-based architecture that naturally blends multiple learners into a precision-ranked cascade of sieves. Each sieve adds labels to the event graph one at a time, and earlier sieves inform later ones through transitive closure. This paper thus describes innovations in both approach and task. We experiment on the densest event graphs to date and show a 14% gain over state-of-the-art.

Subfields: Artificial Intelligence, Computer Networks and Communications

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

In Computer Networks and Communications

Citations in the corpus

ECONET: Effective Continual Pretraining of Language Models for Event Temporal Reasoning (2020, methodology): “In this work, we study two prominent event temporal reasoning tasks as shown in Figure 1: event relation extraction (ERE) (Chambers et al., 2014; Ning et al., 2018; O’Gorman et al., 2016; Mostafazadeh et al., 2016) that predicts temporal relations between a pair of events, and machine reading…”
Domain Knowledge Empowered Structured Neural Net for End-to-End Event Temporal Relation Extraction (2020, unknown): “We use CAEVO (Chambers et al., 2014), a hybrid system of rules and linguistic feature-based MaxEnt classiﬁer, as our feature-based benchmark for TimeBank-Dense.”
A Multi-Axis Annotation Scheme for Event Temporal Relations (2018, methodology): “We can see, though, that the overall performance on MATRES is much better than those in the literature for TempRel extraction, which used to be in the low 50’s (Chambers et al., 2014; Ning et al., 2017).”
Effective Distant Supervision for Temporal Relation Extraction (2020, methodology): “We use CAEVO (Chambers et al., 2014) to detect event pairs and link events to timexes, which include explicit datetimes (January, 1961), relative times (tomorrow) and other natural language indicators (now, until recently).”
Contextualized Word Embeddings Enhanced Event Temporal Relation Extraction for Story Understanding (2019, background): “CAEVO builds a pipeline with ordered sieves.”
TORQUE: A Reading Comprehension Dataset of Temporal Ordering Questions (2020, methodology): “Later works on event and time have largely followed this deﬁnition, e.g., TempEval (Verhagen et al., 2007), TimeBank-Dense (Chambers et al., 2014), RED (O’Gorman et al., 2016), and MATRES (Ning et al., 2018b).”
Joint Event and Temporal Relation Extraction with Shared Representations and Structured Prediction (2019, methodology): “annotated temporal relation corpora with all events and relations fully annotated is reported to be a challenging task as annotators could easily overlook some facts (Bethard et al., 2007; Cassidy et al., 2014; Chambers et al., 2014; Ning et al., 2017), which made both modeling and evaluation extrem”
Joint Reasoning for Temporal and Causal Relations (2018, methodology): “CATENA (Mirza and Tonelli, 2016) extended the multi-sieve framework of CAEVO to extracting both temporal and causal relations and exploited their interaction through post-editing temporal relations based on causal predictions.”
Temporal Common Sense Acquisition with Minimal Supervision (2020, background): “inference, and QA (UzZaman et al., 2013; Chambers et al., 2014; Llorens et al., 2015; Bethard et al., 2016; Leeuwenberg and Moens, 2017; Ning et al., 2018b), rely on understanding time.”
Clinical Temporal Relation Extraction with Probabilistic Soft Logic Regularization and Global Inference (2020, unknown): “(1) CAEVO (Chambers et al. 2014) with a cascade of rule-based classifiers; (2) LSTM-DP (Cheng and Miyao 2017) using LSTM-based network and crosssentence dependency paths; (3) GCL (Meng and Rumshisky 2018) incorporating LSTM-based network with discourselevel contexts; (4) SP-ILP and CTRL, same as…”
Deep Structured Neural Network for Event Temporal Relation Extraction (2019, methodology): “A noteworthy work, CAEVO (Chambers et al., 2014), builds a pipeline with ordered sieves.”