ForecastQA: A Question Answering Challenge for Event Forecasting with Temporal Text Data

  title={ForecastQA: A Question Answering Challenge for Event Forecasting with Temporal Text Data},
  author={Woojeong Jin and Suji Kim and Rahul Khanna and Dong-Ho Lee and Fred Morstatter and A. G. Galstyan and Xiang Ren},
Event forecasting is a challenging, yet important task, as humans seek to constantly plan for the future. Existing automated forecasting studies rely mostly on structured data, such as time-series or event-based knowledge graphs, to help predict future events. In this work, we aim to formulate a task, construct a dataset, and provide benchmarks for developing methods for event forecasting with large volumes of unstructured text data. To simulate the forecasting scenario on temporal news… 
Improving Time Sensitivity for Question Answering over Temporal Knowledge Graphs
A time-sensitive question answering (TSQA) framework that significantly outperforms the previous state of the art on a new benchmark for question answering over temporal KGs, especially achieving a 32% error reduction on complex questions that require multiple steps of reasoning over facts in the temporal KG.
TempoQR: Temporal Question Reasoning over Knowledge Graphs
This paper puts forth a comprehensive embedding-based framework for answering complex questions over TKGs, called TempoQR, which exploits TKG embeddings to ground the question to the specific entities and time scope it refers to and improves accuracy by 25–45 percentage points on complex temporal questions over state-of-the-art approaches.
Unsolved Problems in ML Safety
A new roadmap for ML Safety is provided and four problems ready for research are presented, namely withstanding hazards, identifying hazards), identifying hazards, reducing inherent model hazards, and reducing systemic hazards.
Open Temporal Relation Extraction for Question Answering
This paper decomposes each question into a question event and an open temporal relation (OTR) which is not pre-defined nor with timestamps, and ground the former in the context while sharing the representation of the latter across contexts.
ECONET: Effective Continual Pretraining of Language Models for Event Temporal Reasoning
A continual pre-training approach that equips PTLMs with targeted knowledge about event temporal relations and design self-supervised learning objectives to recover masked-out event and temporal indicators and to discriminate sentences from their corrupted counterparts.


UnifiedQA: Crossing Format Boundaries With a Single QA System
This work uses the latest advances in language modeling to build a single pre-trained QA model, UNIFIEDQA, that performs well across 19 QA datasets spanning 4 diverse formats, and results in a new state of the art on 10 factoid and commonsense question answering datasets.
Dense Passage Retrieval for Open-Domain Question Answering
This work shows that retrieval can be practically implemented using dense representations alone, where embeddings are learned from a small number of questions and passages by a simple dual-encoder framework.
“Going on a vacation” takes longer than “Going for a walk”: A Study of Temporal Commonsense Understanding
It is found that the best current methods used on MCTACO are still far behind human performance, by about 20%, and several directions for improvement are discussed.
SAM-Net: Integrating Event-Level and Chain-Level Attentions to Predict What Happens Next
Self attention mechanism to focus on diverse event segments within the chain and the event chain is represented as a set of event segments, which achieves better results than other state-of-the-art baselines by adopting the evaluation of Multi-Choice Narrative Cloze task.
Answering Complex Open-domain Questions Through Iterative Query Generation
This work presents GoldEn (Gold Entity) Retriever, which iterates between reading context and retrieving more supporting documents to answer open-domain multi-hop questions, and demonstrates that it outperforms the best previously published model despite not using pretrained language models such as BERT.
ALBERT: A Lite BERT for Self-supervised Learning of Language Representations
This work presents two parameter-reduction techniques to lower memory consumption and increase the training speed of BERT, and uses a self-supervised loss that focuses on modeling inter-sentence coherence.
RoBERTa: A Robustly Optimized BERT Pretraining Approach
It is found that BERT was significantly undertrained, and can match or exceed the performance of every model published after it, and the best model achieves state-of-the-art results on GLUE, RACE and SQuAD.
Open Domain Event Extraction Using Neural Latent Variable Models
This work considers open domain event extraction, the task of extracting unconstraint types of events from news clusters, and proposes a novel latent variable neural model, which is scalable to very large corpus.
CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge
This work presents CommonsenseQA: a challenging new dataset for commonsense question answering, which extracts from ConceptNet multiple target concepts that have the same semantic relation to a single source concept.
Deep Contextualized Word Representations
A new type of deep contextualized word representation is introduced that models both complex characteristics of word use and how these uses vary across linguistic contexts, allowing downstream models to mix different types of semi-supervision signals.