arXivDaily arXiv每日学术速递 周一至周五更新
​通知:联邦节假日调休:哥伦布日 / 原住民日(Columbus Day / Indigenous Peoples' Day),今日(13日)无更新。
arXiv 2501.11166cs.CLcs.AIcs.LG

AIMA at SemEval-2024 Task 10: 基于历史的印地语-英语代码混合对话情感识别

AIMA at SemEval-2024 Task 10: History-Based Emotion Recognition in Hindi-English Code-Mixed Conversations

Sharif University of Technology(谢里夫理工大学);Qatar Computing Research Institute(卡塔尔计算研究所)

Mohammad Mahdi Abootorabi, Nona Ghazizadeh, Seyed Arshan Dalili, Alireza Ghahramani Kure, Mahshid Dehghani, Ehsaneddin Asgari

AI总结:

针对SemEval 2024任务10子任务1,提出结合前后上下文及序列信息、并利用印地语-英语翻译流水线的多模型集成方法,实现代码混合对话情感识别并超越基线。

摘要:

在本研究中,我们针对SemEval 2024任务10的子任务1提出了一种解决方案,该子任务专注于代码混合的印地语-英语对话中的情感识别(ERC)。代码混合对话中的ERC带来了独特的挑战,因为现有模型通常在单语数据集上训练,可能无法在代码混合数据上表现良好。为了解决这个问题,我们提出了一系列模型,这些模型结合了当前话语的先前和未来上下文,以及对话的序列信息。为了促进代码混合数据的处理,我们开发了一个印地语-英语翻译流水线,将代码混合对话翻译成英语。我们设计了四种不同的基础模型,每种模型都利用强大的预训练编码器从输入中提取特征,但架构各不相同。通过集成所有这些模型,我们开发了一个最终模型,其性能优于所有其他基线。

英文摘要:

In this study, we introduce a solution to the SemEval 2024 Task 10 on subtask 1, dedicated to Emotion Recognition in Conversation (ERC) in code-mixed Hindi-English conversations. ERC in code-mixed conversations presents unique challenges, as existing models are typically trained on monolingual datasets and may not perform well on code-mixed data. To address this, we propose a series of models that incorporate both the previous and future context of the current utterance, as well as the sequential information of the conversation. To facilitate the processing of code-mixed data, we developed a Hinglish-to-English translation pipeline to translate the code-mixed conversations into English. We designed four different base models, each utilizing powerful pre-trained encoders to extract features from the input but with varying architectures. By ensembling all of these models, we developed a final model that outperforms all other baselines.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑