# Using Emotion Embeddings to Transfer Knowledge Between Emotions, Languages, and Annotation Formats

arXiv: https://arxiv.org/abs/2211.00171 · HTML: https://mlsanity.com/paper/2211.00171/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 35 references that are in the map.
- First posted on arXiv 31 Oct 2022.

## Builds on (most cited first, 18 of 18)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [Cross-lingual Language Model Pretraining](https://mlsanity.com/paper/1901.07291/) (2019)
- [DailyDialog: A Manually Labelled Multi-turn Dialogue Dataset](https://mlsanity.com/paper/1710.03957/) (2017)
- [AraBERT: Transformer-based Model for Arabic Language Understanding](https://mlsanity.com/paper/2003.00104/) (2020)
- [BERTweet: A pre-trained language model for English Tweets](https://mlsanity.com/paper/2005.10200/) (2020)
- [Benchmarking Zero-shot Text Classification: Datasets, Evaluation and Entailment Approach](https://mlsanity.com/paper/1909.00161/) (2019)
- [Using millions of emoji occurrences to learn any-domain representations for detecting sentiment, emotion and sarcasm](https://mlsanity.com/paper/1708.00524/) (2017)
- [XLM-T: Multilingual Language Models in Twitter for Sentiment Analysis and Beyond](https://mlsanity.com/paper/2104.12250/) (2021)
- [NTUA-SLP at SemEval-2018 Task 1: Predicting Affective Content in Tweets with Deep Attentive RNNs and Transfer Learning](https://mlsanity.com/paper/1804.06658/) (2018)
- [RoBERTuito: a pre-trained language model for social media text in Spanish](https://mlsanity.com/paper/2111.09453/) (2021)
- [Leveraging Label Correlations in a Multi-label Setting: A Case Study in Emotion](https://mlsanity.com/paper/2210.15842/) (2022)
- [Natural Language Inference Prompts for Zero-shot Emotion Classification in Text across Corpora](https://mlsanity.com/paper/2209.06701/) (2022)
- [Joint Binary Neural Network for Multi-label Learning with Applications to Emotion Classification](https://mlsanity.com/paper/1802.00891/) (2018)
- [EmoGraph: Capturing Emotion Correlations using Graph Networks](https://mlsanity.com/paper/2008.09378/) (2020)
- [BERTweetFR : Domain Adaptation of Pre-Trained Language Models for French Tweets](https://mlsanity.com/paper/2109.10234/) (2021)
- [Emotion Regulation and Dynamics of Moral Concerns During the Early COVID-19 Pandemic](https://mlsanity.com/paper/2203.03608/) (2022)

## Built on by (most cited first, 7 of 7)

- [Whose Emotions and Moral Sentiments Do Language Models Reflect?](https://mlsanity.com/paper/2402.11114/) (2024)
- [Anger Breeds Controversy: Analyzing Controversy and Emotions on Reddit](https://mlsanity.com/paper/2212.00339/) (2022)
- [Improving and Assessing the Fidelity of Large Language Models Alignment to Online Communities](https://mlsanity.com/paper/2408.09366/) (2024)
- [Generating Fearful Images: Investigating Potential Emotional Biases in Image-Generation Models](https://mlsanity.com/paper/2411.05985/) (2024)
- [Contrastive Distillation of Emotion Knowledge from LLMs for Zero-Shot Emotion Recognition](https://mlsanity.com/paper/2505.18040/) (2025)
- [Semantic F1 Scores: Fair Evaluation Under Fuzzy Class Boundaries](https://mlsanity.com/paper/2509.21633/) (2025)
- [A Comparative Study on Affective Cues in Text Embeddings Across Psychological Emotion Theories](https://mlsanity.com/paper/2606.29068/) (2026)

## Abstract (arXiv)

The need for emotional inference from text continues to diversify as more and more disciplines integrate emotions into their theories and applications. These needs include inferring different emotion types, handling multiple languages, and different annotation formats. A shared model between different configurations would enable the sharing of knowledge and a decrease in training costs, and would simplify the process of deploying emotion recognition models in novel environments. In this work, we study how we can build a single model that can transition between these different configurations by leveraging multilingual models and Demux, a transformer-based model whose input includes the emotions of interest, enabling us to dynamically change the emotions predicted by the model. Demux also produces emotion embeddings, and performing operations on them allows us to transition to clusters of emotions by pooling the embeddings of each cluster. We show that Demux can simultaneously transfer knowledge in a zero-shot manner to a new language, to a novel annotation format and to unseen emotions. Code is available at https://github.com/gchochla/Demux-MEmo .