EMNLP Findings2020
Improving Event Duration Prediction via Time-aware Pre-training
Zonglin Yang, Xinya Du, Alexander M. Rush, Claire Cardie
Abstract
End-to-end models in NLP rarely encode external world knowledge about length of time. We introduce two effective models for duration prediction, which incorporate external knowledge by reading temporal-related news sentences (time-aware pre-training). Specifically, one model predicts the range/unit where the duration value falls in (R-pred); and the other predicts the exact duration value E-pred. Our best model -- E-pred, substantially outperforms previous work, and captures duration information more accurately than R-pred. We also demonstrate our models are capable of duration prediction in the unsupervised setting, outperforming the baselines.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
6 top-tier papers have cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
- Mitigating Reporting Bias in Semi-supervised Temporal Commonsense Inference with Probabilistic Soft Logic cites Temporal Common Sense Acquisition with Minimal Supervision
- Mitigating Temporal Misalignment by Discarding Outdated Facts cites Temporal Common Sense Acquisition with Minimal Supervision
- Self-Supervised Logic Induction for Explainable Fuzzy Temporal Commonsense Reasoning cites Temporal Common Sense Acquisition with Minimal Supervision
Cited by top-tier papers6
- Mitigating Reporting Bias in Semi-supervised Temporal Commonsense Inference with Probabilistic Soft LogicBibo Cai, Xiao Ding, Bowen Chen, Li Du et al.AAAI 2022 · 12 citations
- Self-Supervised Logic Induction for Explainable Fuzzy Temporal Commonsense ReasoningBibo Cai, Xiao Ding, Zhouhao Sun, Bing Qin et al.AAAI 2023 · 11 citations
- Mitigating Temporal Misalignment by Discarding Outdated FactsMichael J. Q. Zhang, Eunsol ChoiEMNLP 2023 · 6 citations
- It's High Time: A Survey of Temporal Question AnsweringBhawna Piryani, Abdelrahman Abdallah, Jamshid Mozafari, Avishek Anand et al.ACL 2026 · 6 citations
- MOOSE-Chem: Large Language Models for Rediscovering Unseen Chemistry Scientific HypothesesZonglin Yang, Wanhao Liu, Ben Gao, Tong Xie et al.ICLR 2025 · 2 citations
Builds on2
- Temporal Common Sense Acquisition with Minimal SupervisionBen Zhou, Qiang Ning, Daniel Khashabi, Dan RothACL 2020 · 76 citations
- Pre-training Is (Almost) All You Need: An Application to Commonsense ReasoningAlexandre Tamborrino, Nicola Pellicanò, Baptiste Pannier, Pascal Voitot et al.ACL 2020 · 30 citations
Related papers
- BiTimeBERT: Extending Pre-Trained Language Representations with Bi-Temporal InformationJiexin Wang, Adam Jatowt, Masatoshi Yoshikawa, Yi CaiSIGIR 2023 · 15 citations
- Temporal Reasoning in Natural Language InferenceSiddharth Vashishtha, Adam Poliak, Yash Kumar Lal, Benjamin Van Durme et al.EMNLP Findings 2020 · 31 citations
- JAKET: Joint Pre-training of Knowledge Graph and Language UnderstandingDonghan Yu, Chenguang Zhu, Yiming Yang, Michael ZengAAAI 2022 · 171 citations
- PUNR: Pre-training with User Behavior Modeling for News RecommendationGuangyuan Ma, Hongtao Liu, Xing Wu, Wanhui Qian et al.EMNLP Findings 2023 · 5 citations
- Generating Coherent Narratives by Learning Dynamic and Discrete Entity States with a Contrastive FrameworkJian Guan, Zhenyu Yang, Rongsheng Zhang, Zhipeng Hu et al.AAAI 2023 · 11 citations
