mlsanityData to 7 Oct 2026

Using millions of emoji occurrences to learn any-domain representations for detecting sentiment, emotion and sarcasm

arXiv 1708.00524 ↗First posted 1 Aug 2017stat.ML cs.LGcited by 117

Not on the map. Why?

Abstract (arXiv)

NLP tasks are often limited by scarcity of manually annotated data. In social media sentiment analysis and related tasks, researchers have therefore used binarized emoticons and specific hashtags as forms of distant supervision. Our paper shows that by extending the distant supervision to a more diverse set of noisy labels, the models can learn richer representations. Through emoji prediction on a dataset of 1246 million tweets containing one of 64 common emojis we obtain state-of-the-art performance on 8 benchmark datasets within sentiment, emotion and sarcasm detection using a single pretrained model. Our analyses confirm that the diversity of our emotional labels yield a performance improvement over previous distant supervision approaches.

Key facts. Cited by 117 papers in the map (data through 7 Oct 2026). First posted on arXiv 1 Aug 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2019: 1; 2020: 1; 2021: 38; 2022: 27; 2023: 21; 2024: 14; 2025: 11; 2026: 4.

117 papers.

Sort:

By area:

  1. 2019Way Off-Policy Batch Deep Reinforcement Learning of Implicit Human Preferences in Dialogcited by 283
  2. 2021IconQA: A New Benchmark for Abstract Diagram Understanding and Visual Language Reasoningcited by 262
  3. 2023emotion2vec: Self-Supervised Pre-Training for Speech Emotion Representationcited by 196
  4. 2023Stop Uploading Test Data in Plain Text: Practical Strategies for Mitigating Data Contamination by Evaluation Benchmarkscited by 139
  5. 2020Human-centric Dialog Training via Offline Reinforcement Learningcited by 92
  6. 2021Hatemoji: A Test Suite and Adversarially-Generated Dataset for Benchmarking and Detecting Emoji-based Hatecited by 50
  7. 2021Diverse Adversaries for Mitigating Bias in Trainingcited by 48
  8. 2022Dimensional Modeling of Emotions in Text with Appraisal Theories: Corpus Creation, Annotation Reliability, and Predictioncited by 45
  9. 2021SpanEmo: Casting Multi-label Emotion Classification as Span-predictioncited by 37
  10. 2021Balancing out Bias: Achieving Fairness Through Balanced Trainingcited by 34
  11. 2021Contrastive Learning for Fair Representationscited by 34
  12. 2022Gold Doesn't Always Glitter: Spectral Removal of Linear and Nonlinear Guarded Attribute Informationcited by 33
38 more
  1. 2021Refining Language Models with Compositional Explanationscited by 27
  2. 2022Towards Multi-Modal Sarcasm Detection via Hierarchical Congruity Modeling with Knowledge Enhancementcited by 26
  3. 2021Nice perfume. How long did you marinate in it? Multimodal Sarcasm Explanationcited by 25
  4. 2023Shielded Representations: Protecting Sensitive Attributes Through Iterative Gradient-Based Projectioncited by 25
  5. 2021Self-Supervised Euphemism Detection and Identification for Content Moderationcited by 24
  6. 2022Optimising Equal Opportunity Fairness in Model Trainingcited by 22
  7. 2023Multi-source Semantic Graph-based Multimodal Sarcasm Explanation Generationcited by 19
  8. 2022Identifying Moments of Change from Longitudinal User Textcited by 17
  9. 2022Learning Fair Representations via Rate-Distortion Maximizationcited by 15
  10. 2021A Dual-Channel Framework for Sarcasm Recognition by Detecting Sentiment Conflictcited by 14
  11. 2022Empathetic Conversational Systems: A Review of Current Advances, Gaps, and Opportunitiescited by 14
  12. 2022Leveraging Label Correlations in a Multi-label Setting: A Case Study in Emotioncited by 14
  13. 2023ShuttleSet: A Human-Annotated Stroke-Level Singles Dataset for Badminton Tactical Analysiscited by 13
  14. 2021Exploring the Long Short-Term Dependencies to Infer Shot Influence in Badminton Matchescited by 10
  15. 2021A Little Pretraining Goes a Long Way: A Case Study on Dependency Parsing Task for Low-resource Morphologically Rich Languagescited by 9
  16. 2021Modulating Language Models with Emotionscited by 9
  17. 2022Fair NLP Models with Differentially Private Text Encoderscited by 9
  18. 2023StockEmotions: Discover Investor Emotions for Financial Sentiment Analysis and Multivariate Time Seriescited by 9
  19. 2022Towards Equal Opportunity Fairness through Adversarial Learningcited by 8
  20. 2022FairGrad: Fairness Aware Gradient Descentcited by 8
  21. 2021CheerBots: Chatbots toward Empathy and Emotionusing Reinforcement Learningcited by 7
  22. 2022Using Emotion Embeddings to Transfer Knowledge Between Emotions, Languages, and Annotation Formatscited by 7
  23. 2023A Comprehensive Survey on Affective Computing; Challenges, Trends, Applications, and Future Directionscited by 7
  24. 2023Where are We in Event-centric Emotion Analysis? Bridging Emotion Role Labeling and Appraisal-based Approachescited by 7
  25. 2024Emojis Decoded: Leveraging ChatGPT for Enhanced Understanding in Social Media Communicationscited by 7
  26. 2021Latent-Optimized Adversarial Neural Transfer for Sarcasm Detectioncited by 6
  27. 2021Offensive Language Identification in Low-resourced Code-mixed Dravidian languages using Pseudo-labelingcited by 6
  28. 2023Erasure of Unaligned Attributes from Neural Representationscited by 6
  29. 2024Sentiment-enhanced Graph-based Sarcasm Explanation in Dialoguecited by 6
  30. 2025Emergence of Hierarchical Emotion Organization in Large Language Modelscited by 6
  31. 2021On the Impact of Word Error Rate on Acoustic-Linguistic Speech Emotion Recognition: An Update for the Deep Learning Eracited by 5
  32. 2021Improving Social Meaning Detection with Pragmatic Masking and Surrogate Fine-Tuningcited by 5
  33. 2021From Theories on Styles to their Transfer in Text: Bridging the Gap with a Hierarchical Surveycited by 5
  34. 2022A Federated Approach to Predicting Emojis in Hindi Tweetscited by 5
  35. 2021Embodying Pre-Trained Word Embeddings Through Robot Actionscited by 4
  36. 2021Nora: The Well-Being Coachcited by 4
  37. 2021Transfer-based adaptive tree for multimodal sentiment analysis based on user latent aspectscited by 4
  38. 2021DYPLODOC: Dynamic Plots for Document Classificationcited by 4

50 most cited of 117 All 117 (JSON)