資訊：大學(xué)研究人員開(kāi)發(fā)出檢測(cè)諷刺語(yǔ)態(tài)的AI平臺(tái)

University researchers develop AI platform to detect sarcasm

大學(xué)研究人員開(kāi)發(fā)出檢測(cè)諷刺語(yǔ)態(tài)的AI平臺(tái)


Researchers from the Netherlands have developed an AI platform capable of recognising sarcasm. The project was presented at a meeting of the Acoustical Society of America and the Canadian Acoustical Association in Ottawa. Using video clips and text from American sitcoms such as ‘Friends’ and ‘The Big Bang Theory,’ the researchers trained a neural network with the Multimodal Sarcasm Detection Dataset (MUStARD), previously annotated by another research team from the US and Singapore.

荷蘭研究人員開(kāi)發(fā)了一個(gè)能夠識(shí)別諷刺語(yǔ)態(tài)的AI平臺(tái)。該項(xiàng)目在渥太華舉行的美國(guó)聲學(xué)協(xié)會(huì)和加拿大聲學(xué)協(xié)會(huì)會(huì)議上亮相。研究人員利用《老友記》和《生活大爆炸》等美國(guó)情景喜劇中的視頻剪輯和文本，使用多模態(tài)諷刺檢測(cè)數(shù)據(jù)集 (MUStARD) 訓(xùn)練神經(jīng)網(wǎng)絡(luò)，該數(shù)據(jù)集此前由來(lái)自美國(guó)和新加坡的另一個(gè)研究團(tuán)隊(duì)注釋。

After being trained on this data, the AI model successfully detected sarcasm in unlabeled exchanges about 75% of the time. Further improvements using synthetic data have reportedly enhanced this accuracy, though these findings are yet to be published. Notable scenes used for training included moments from ‘The Big Bang Theory’ and ‘Friends’ that exemplified sarcastic interactions.

經(jīng)過(guò)這些數(shù)據(jù)的訓(xùn)練后，AI模型在大約75%的情況下成功檢測(cè)到交流中的諷刺。據(jù)報(bào)道，使用合成數(shù)據(jù)進(jìn)一步提高了這種準(zhǔn)確性，盡管這些發(fā)現(xiàn)尚未發(fā)表。用于訓(xùn)練的場(chǎng)景有《生活大爆炸》和《老友記》中諷刺互動(dòng)的片段。

The research team at the University of Groningen aims to advance their sarcasm detection capabilities. By incorporating visual cues such as facial expressions, they aim to refine the AI’s ability to detect sarcasm more accurately. The project could significantly improve AI assistants’ interactions by enabling them to understand negative or hostile tones in human speech.

格羅寧根大學(xué)的研究團(tuán)隊(duì)旨在提高AI的諷刺檢測(cè)能力。通過(guò)結(jié)合面部表情等視覺(jué)線(xiàn)索，他們的目標(biāo)是提高人工智能更準(zhǔn)確地檢測(cè)諷刺的能力。該項(xiàng)目可以讓人工智能助理理解人類(lèi)言語(yǔ)中的負(fù)面或敵意語(yǔ)氣，從而顯著改善他們的互動(dòng)。

Sarcasm generally takes the form of an ironic remark, often rooted in humour, that is intended to mock or satirise something. When a speaker is being sarcastic, they say something different than what they actually mean, and that’s why it is hard for a large language machine to detect such nuances in someone’s speech.

諷刺通常植根于幽默，旨在嘲笑或諷刺某些事物。當(dāng)說(shuō)話(huà)者諷刺時(shí)，他們所說(shuō)的內(nèi)容與實(shí)際意思不同，這就是為什么大型語(yǔ)言機(jī)器很難檢測(cè)某人講話(huà)中的細(xì)微差別。

The project aligns with similar research initiatives, such as those by the US Department of Defense’s DARPA, which developed an AI model for detecting sarcasm in text. The success of these projects is a significant achievement since it underscores the importance of understanding sarcasm in human communication, which could enhance the development of more nuanced and compelling AI systems.

該項(xiàng)目與類(lèi)似的研究計(jì)劃相一致，例如美國(guó)國(guó)防部DARPA的研究計(jì)劃，該計(jì)劃開(kāi)發(fā)了一種用于檢測(cè)文本中諷刺內(nèi)容的人工智能模型。這些項(xiàng)目的成功是一項(xiàng)重大成就，因?yàn)樗鼜?qiáng)調(diào)了理解人類(lèi)交流中諷刺的重要性，這可以促進(jìn)更細(xì)致、更引人注目的人工智能系統(tǒng)的開(kāi)發(fā)。