# RTHN: A RNN-Transformer Hierarchical Network for Emotion Cause Extraction

arXiv: https://arxiv.org/abs/1906.01236 · HTML: https://mlsanity.com/paper/1906.01236/

## Key facts

- Cited by 25 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 4 Jun 2019.

## Built on by (most cited first, 25 of 25)

- [One Chatbot Per Person: Creating Personalized Chatbots based on Implicit User Profiles](https://mlsanity.com/paper/2108.09355/) (2021)
- [Multimodal Emotion-Cause Pair Extraction in Conversations](https://mlsanity.com/paper/2110.08020/) (2021)
- [Position Bias Mitigation: A Knowledge-Aware Graph Model for Emotion Cause Extraction](https://mlsanity.com/paper/2106.03518/) (2021)
- [Knowledge-Bridged Causal Interaction Network for Causal Emotion Entailment](https://mlsanity.com/paper/2212.02995/) (2022)
- [An End-to-End Network for Emotion-Cause Pair Extraction](https://mlsanity.com/paper/2103.01544/) (2021)
- [Neutral Utterances are Also Causes: Enhancing Conversational Causal Emotion Entailment with Social Commonsense Knowledge](https://mlsanity.com/paper/2205.00759/) (2022)
- [SemEval-2024 Task 3: Multimodal Emotion Cause Analysis in Conversations](https://mlsanity.com/paper/2405.13049/) (2024)
- [Empathetic Response Generation through Graph-based Multi-hop Reasoning on Emotional Causality](https://mlsanity.com/paper/2110.04614/) (2021)
- [Multi-Task Learning and Adapted Knowledge Models for Emotion-Cause Extraction](https://mlsanity.com/paper/2106.09790/) (2021)
- [TSAM: A Two-Stream Attention Model for Causal Emotion Entailment](https://mlsanity.com/paper/2203.00819/) (2022)
- [Emotion Stimulus Detection in German News Headlines](https://mlsanity.com/paper/2107.12920/) (2021)
- [Emotion Prediction Oriented method with Multiple Supervisions for Emotion-Cause Pair Extraction](https://mlsanity.com/paper/2302.12417/) (2023)
- [Emotion Flip Reasoning in Multiparty Conversations](https://mlsanity.com/paper/2306.13959/) (2023)
- [A Dual-Questioning Attention Network for Emotion-Cause Pair Extraction with Context Awareness](https://mlsanity.com/paper/2104.07221/) (2021)
- [Multi-Task Learning Framework for Extracting Emotion Cause Span and Entailment in Conversations](https://mlsanity.com/paper/2211.03742/) (2022)
- [PAGE: A Position-Aware Graph-Based Model for Emotion Cause Entailment in Conversation](https://mlsanity.com/paper/2303.01795/) (2023)
- [Hierarchical RNNs-Based Transformers MADDPG for Mixed Cooperative-Competitive Environments](https://mlsanity.com/paper/2105.04888/) (2021)
- [Pair-Based Joint Encoding with Relational Graph Convolutional Networks for Emotion-Cause Pair Extraction](https://mlsanity.com/paper/2212.01844/) (2022)
- [ConTrip: Consensus Sentiment review Analysis and Platform ratings in a single score](https://mlsanity.com/paper/2201.02113/) (2022)
- [A Multi-turn Machine Reading Comprehension Framework with Rethink Mechanism for Emotion-Cause Pair Extraction](https://mlsanity.com/paper/2209.07972/) (2022)
- [Recognizing Conditional Causal Relationships about Emotions and Their Corresponding Conditions](https://mlsanity.com/paper/2311.16579/) (2023)
- [Emotion-cause pair extraction method based on multi-granularity information and multi-module interaction](https://mlsanity.com/paper/2404.06812/) (2024)
- [Generative Emotion Cause Explanation in Multimodal Conversations](https://mlsanity.com/paper/2411.02430/) (2024)
- [Acquiring Common Chinese Emotional Events Using Large Language Model](https://mlsanity.com/paper/2511.04989/) (2025)
- [Building LLM Agent Systems the Deep Learning Way: From Modular Design to Architecture Search](https://mlsanity.com/paper/2610.04961/) (2026)

## Abstract (arXiv)

The emotion cause extraction (ECE) task aims at discovering the potential causes behind a certain emotion expression in a document. Techniques including rule-based methods, traditional machine learning methods and deep neural networks have been proposed to solve this task. However, most of the previous work considered ECE as a set of independent clause classification problems and ignored the relations between multiple clauses in a document. In this work, we propose a joint emotion cause extraction framework, named RNN-Transformer Hierarchical Network (RTHN), to encode and classify multiple clauses synchronously. RTHN is composed of a lower word-level encoder based on RNNs to encode multiple words in each clause, and an upper clause-level encoder based on Transformer to learn the correlation between multiple clauses in a document. We furthermore propose ways to encode the relative position and global predication information into Transformer that can capture the causality between clauses and make RTHN more efficient. We finally achieve the best performance among 12 compared systems and improve the F1 score of the state-of-the-art from 72.69% to 76.77%.