An Iterative Emotion Interaction Network for Emotion Recognition in Conversations

Lu, Xin; Zhao, Yanyan; Wu, Yang; Tian, Yijian; Chen, Huipeng; Qin, Bing

doi:10.18653/v1/2020.coling-main.360

Cited by 28 publications

(11 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A substantial number of approaches are heavily based on RNNs (e.g., GRU) to model the sequence (Jiao et al, 2019), (Lu et al, 2020). The biggest problem with this is that it inherently decouples the word embedding extraction and the sequence modeling, whereas BERT-like models tackle them at once, often leading to a better performance.…”

Section: Related Workmentioning

confidence: 99%

“…MELD IEMOCAP BERT+MTL (Li et al, 2020b) 61.90 -BiERU-lc (Li et al, 2020c) 60.84 64.65 DialogueGCN (Ghosal et al, 2019) 58.1 64.18 RGAT (Ishiwatari et al, 2020) 60.91 65.22 CESTa 58.36 67.1 VHRED (Hazarika et al, 2021) -58.6 SumAggGIN (Sheng et al, 2020) 58.45 66.61 COSMIC (Ghosal et al, 2020) 65.21 65.28 KET (Zhong et al, 2019b) 58.18 59.56 BiF-AGRU (Jiao et al, 2019) 58.1 63.5 Iterative (Lu et al, 2020) 60.72 64.37 HiTrans (Li et al, 2020a) 61.94 64.5 DialogXL 62 EmoBERTa shows very good results: max weighted f 1 scores (%) of 65.61 (MELD) and 68.57 (IEMOCAP) respectively and above the best reported SOTA, especially considering that no modifications were made to the original RoBERTa model architecture. We also trained a model without the speaker names prepended, which drops the performance: weighted f 1 scores (%) of 65.07 (MELD) and 64.02 (IEMOCAP) respectively, pro-viding evidence that encoding the speaker information helps.…”

Section: Modelmentioning

confidence: 99%

See 1 more Smart Citation

EmoBERTa: Speaker-Aware Emotion Recognition in Conversation with RoBERTa

Kim¹,

Vossen²

2021

Preprint

View full text Add to dashboard Cite

We present EmoBERTa: Speaker-Aware Emotion Recognition in Conversation with RoBERTa, a simple yet expressive scheme of solving the ERC (emotion recognition in conversation) task. By simply prepending speaker names to utterances and inserting separation tokens between the utterances in a dialogue, EmoBERTa can learn intra-and inter-speaker states and context to predict the emotion of a current speaker, in an end-to-end manner. Our experiments show that we reach a new state of the art on the two popular ERC datasets using a basic and straight-forward approach.We've open sourced our code and models at https://github.com/tae898/erc.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Modelmentioning

confidence: 99%

EmoBERTa: Speaker-Aware Emotion Recognition in Conversation with RoBERTa

Kim¹,

Vossen²

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…modeled ERC as sequence tagging to learn the emotional consistency. Lu et al (2020) proposed an iterative emotion interaction network to explicitly model the emotion interaction.…”

Section: Emotion Recognition In Conversationsmentioning

confidence: 99%

“…2) The integration of emotional clues. Many works Ghosal et al, 2019;Lu et al, 2020) usually use the attention mechanism to integrate encoded emotional clues, ignoring their intrinsic semantic order. It would lose logical relationships between clues, making it difficult to capture key factors that trigger emotions.…”

Section: Introductionmentioning

confidence: 99%

DialogueCRN: Contextual Reasoning Networks for Emotion Recognition in Conversations

Hu¹,

Wei²,

Huai³

2021

Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Confer

View full text Add to dashboard Cite

Emotion Recognition in Conversations (ERC) has gained increasing attention for developing empathetic machines. Recently, many approaches have been devoted to perceiving conversational context by deep learning models. However, these approaches are insufficient in understanding the context due to lacking the ability to extract and integrate emotional clues. In this work, we propose novel Contextual Reasoning Networks (DialogueCRN) to fully understand the conversational context from a cognitive perspective. Inspired by the Cognitive Theory of Emotion, we design multiturn reasoning modules to extract and integrate emotional clues. The reasoning module iteratively performs an intuitive retrieving process and a conscious reasoning process, which imitates human unique cognitive thinking. Extensive experiments on three public benchmark datasets demonstrate the effectiveness and superiority of the proposed model.

show abstract

“…IEIN (Lu et al, 2020): IEIN uses predicted emotion labels instead of gold labels and designs a loss to constrain the prediction of each iteration.…”

Section: Baselines and State Of The Artmentioning

confidence: 99%

Knowledge-Interactive Network with Sentiment Polarity Intensity-Aware Multi-Task Learning for Emotion Recognition in Conversations

Xie¹,

Yang²,

Sun³

et al. 2021

Findings of the Association for Computational Linguistics: EMNLP 2021

View full text Add to dashboard Cite

Emotion Recognition in Conversation (ERC) has gained much attention from the NLP community recently. Some models concentrate on leveraging commonsense knowledge or multitask learning to help complicated emotional reasoning. However, these models neglect direct utterance-knowledge interaction. In addition, these models utilize emotion-indirect auxiliary tasks, which provide limited affective information for the ERC task. To address the above issues, we propose a Knowledge-Interactive Network with sentiment polarity intensity-aware multi-task learning, namely KI-Net, which leverages both commonsense knowledge and sentiment lexicon to augment semantic information. Specifically, we use a self-matching module for internal utterance-knowledge interaction. Considering correlations with the ERC task, a phrase-level Sentiment Polarity Intensity Prediction (SPIP) task is devised as an auxiliary task. Experiments show that all knowledge introduction, self-matching and SPIP modules improve the model performance respectively on three datasets. Moreover, our KI-Net model shows 1.04% performance improvement over the state-of-the-art model on the IEMOCAP dataset.

show abstract

An Iterative Emotion Interaction Network for Emotion Recognition in Conversations

Cited by 28 publications

References 17 publications

EmoBERTa: Speaker-Aware Emotion Recognition in Conversation with RoBERTa

EmoBERTa: Speaker-Aware Emotion Recognition in Conversation with RoBERTa

DialogueCRN: Contextual Reasoning Networks for Emotion Recognition in Conversations

Knowledge-Interactive Network with Sentiment Polarity Intensity-Aware Multi-Task Learning for Emotion Recognition in Conversations

Contact Info

Product

Resources

About