mlsanityData to 7 Oct 2026

Uncovering Main Causalities for Long-tailed Information Extraction

arXiv 2109.05213 ↗First posted 11 Sep 2021cs.CLcited by 21

Not on the map. Why?

Abstract (arXiv)

Information Extraction (IE) aims to extract structural information from unstructured texts. In practice, long-tailed distributions caused by the selection bias of a dataset, may lead to incorrect correlations, also known as spurious correlations, between entities and labels in the conventional likelihood models. This motivates us to propose counterfactual IE (CFIE), a novel framework that aims to uncover the main causalities behind data in the view of causal inference. Specifically, 1) we first introduce a unified structural causal model (SCM) for various IE tasks, describing the relationships among variables; 2) with our SCM, we then generate counterfactuals based on an explicit language structure to better calculate the direct causal effect during the inference stage; 3) we further propose a novel debiasing approach to yield more robust predictions. Experiments on three IE tasks across five public datasets show the effectiveness of our CFIE model in mitigating the spurious correlation issues.

Key facts. Cited by 21 papers in the map (data through 7 Oct 2026). Builds on 51 of its 80 references that are in the map. First posted on arXiv 11 Sep 2021.

Builds on

51 of 80 references.

References · 51

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2016V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentationcited by 2,284
  3. 2019Decoupling Representation and Classifier for Long-Tailed Recognitioncited by 806
  4. 2019Large-Scale Long-Tailed Recognition in an Open Worldcited by 704
  5. 2018Exploring the Limits of Weakly Supervised Pretrainingcited by 604
  6. 2016Neural Architectures for Named Entity Recognitioncited by 548
  7. 2019ScanRefer: 3D Object Localization in RGB-D Scans using Natural Languagecited by 446
  8. 2020Unbiased Scene Graph Generation from Biased Trainingcited by 355
  9. 2016End-to-end Sequence Labeling via Bi-directional LSTM-CNNs-CRFcited by 341
  10. 2020Counterfactual VQA: A Cause-Effect Look at Language Biascited by 266
  11. 2020Long-Tailed Classification by Keeping the Good and Removing the Bad Momentum Causal Effectcited by 256
  12. 2020Causal Intervention for Weakly-Supervised Semantic Segmentationcited by 229
38 more
  1. 2020Retrieve, Program, Repeat: Complex Knowledge Base Question Answering via Alternate Meta-learningcited by 216
  2. 2020Scan2Cap: Context-aware Dense Captioning in RGB-D Scanscited by 215
  3. 2015Named Entity Recognition with Bidirectional LSTM-CNNscited by 201
  4. 2018Graph Convolution over Pruned Dependency Trees Improves Relation Extractioncited by 146
  5. 2021Causal Intervention for Leveraging Popularity Bias in Recommendationcited by 116
  6. 2020Interventional Few-Shot Learningcited by 112
  7. 2020The Devil is in Classification: A Simple Framework for Long-tail Object Detection and Instance Segmentationcited by 108
  8. 2021Context-aware Biaffine Localizing Network for Temporal Sentence Groundingcited by 107
  9. 2020MAVEN: A Massive General Domain Event Detection Datasetcited by 103
  10. 2017Cross-Sentence N-ary Relation Extraction with Graph LSTMscited by 101
  11. 2020Reasoning with Latent Structure Refinement for Document-Level Relation Extractioncited by 97
  12. 2020Model-Agnostic Counterfactual Reasoning for Eliminating Popularity Bias in Recommender Systemcited by 94
  13. 2021Interventional Video Grounding with Dual Contrastive Learningcited by 94
  14. 2021SUTD-TrafficQA: A Question Answering Benchmark and an Efficient Network for Video Reasoning over Traffic Eventscited by 84
  15. 2021Deconfounded Video Moment Retrieval with Causal Interventioncited by 84
  16. 2021Document-level Relation Extraction as Semantic Segmentationcited by 80
  17. 2020Learning from Context or Names? An Empirical Study on Neural Relation Extractioncited by 75
  18. 2020Robustness to Spurious Correlations in Text Classification via Automatically Generated Counterfactualscited by 67
  19. 2017Zero-Shot Transfer Learning for Event Extractioncited by 66
  20. 2020Deconfounded Image Captioning: A Causal Retrospectcited by 65
  21. 2016Distant Supervision for Relation Extraction beyond the Sentence Boundarycited by 62
  22. 2020Robustness to Spurious Correlations via Human Annotationscited by 62
  23. 2019Long-tail Relation Extraction via Knowledge Graph Embeddings and Graph Convolution Networkscited by 61
  24. 2021Video Corpus Moment Retrieval with Contrastive Learningcited by 60
  25. 2021Counterfactual Explainable Recommendationcited by 60
  26. 2020Identifying Spurious Correlations for Robust Text Classificationcited by 59
  27. 2020Identifying Causal-Effect Inference Failure with Uncertainty-Aware Modelscited by 57
  28. 2021Natural Language Video Localization: A Revisit in Span-based Question Answering Frameworkcited by 55
  29. 2020Clicks can be Cheating: Counterfactual Recommendation for Mitigating Clickbait Issuecited by 48
  30. 2020Contrastive Triple Extraction with Generative Transformercited by 44
  31. 2021PRGC: Potential Relation and Global Correspondence Based Joint Relational Triple Extractioncited by 40
  32. 2020Few-shot Relation Extraction via Bayesian Meta-learning on Relation Graphscited by 39
  33. 2019Dependency-Guided LSTM-CRF for Named Entity Recognitioncited by 28
  34. 2018N-ary Relation Extraction using Graph State LSTMcited by 26
  35. 2020Should Graph Convolution Trust Neighbors? A Simple Causal Inference Methodcited by 25
  36. 2021Semantic Representation for Dialogue Modelingcited by 24
  37. 2021Better Feature Integration for Named Entity Recognitioncited by 22
  38. 2021Empowering Language Understanding with Counterfactual Reasoningcited by 13

50 most cited of 51 All 51 (JSON)

Built on by

Cited by year

Citing papers per year: 2022: 7; 2023: 9; 2024: 5.

21 papers.

Sort:

By area:

  1. 2022Text and Patterns: For Effective Chain of Thought, It Takes Two to Tangocited by 128
  2. 2023Prompting Large Language Models for Counterfactual Generation: An Empirical Studycited by 36
  3. 2022Should We Rely on Entity Mentions for Relation Extraction? Debiasing Relation Extraction with Counterfactual Analysiscited by 30
  4. 2023A Causal View of Entity Bias in (Large) Language Modelscited by 21
  5. 2022A Survey of Methods for Addressing Class Imbalance in Deep-Learning Based Natural Language Processingcited by 20
  6. 2024Joint Vision-Language Social Bias Removal for CLIPcited by 17
  7. 2022Extracting or Guessing? Improving Faithfulness of Event Temporal Relation Extractioncited by 15
  8. 2023Preserving Commonsense Knowledge from Pre-trained Language Models via Causal Inferencecited by 13
  9. 2024Controlling Risk of Retrieval-augmented Generation: A Counterfactual Prompting Frameworkcited by 12
  10. 2022Distilling Causal Effect from Miscellaneous Other-Class for Continual Named Entity Recognitioncited by 10
  11. 2023Think Rationally about What You See: Continuous Rationale Extraction for Relation Extractioncited by 6
  12. 2023Towards Heterogeneous Long-tailed Learning: Benchmarking, Metrics, and Toolboxcited by 6
9 more
  1. 2023Adaptive Visual Scene Understanding: Incremental Scene Graph Generationcited by 6
  2. 2022Towards Realistic Low-resource Relation Extraction: A Benchmark with Empirical Baseline Studycited by 5
  3. 2023How Fragile is Relation Extraction under Entity Replacements?cited by 5
  4. 2024A Unified Causal View of Instruction Tuningcited by 1
  5. 2024Balancing the Causal Effects in Class-Incremental Learningcited by 1
  6. 2022Towards Generalized Open Information Extractioncited by 0
  7. 2023ReMask: A Robust Information-Masking Approach for Domain Counterfactual Generationcited by 0
  8. 2023Inducing Causal Structure for Abstractive Text Summarizationcited by 0
  9. 2024Exploring Description-Augmented Dataless Intent Classificationcited by 0