في هذا العمل، نركز على سيناريو عددا أقل تحديا للكشف عن قلة الرصاص حيث يكون العديد من النوايا المحبوسة بشكل جيد ومشبه بشكل صحيح.نقدم مخطط اكتشاف عديدي بسيطة ولكنه فعالة من القلة عبر التدريب المسبق والضبط الناعم الصنع.على وجه التحديد، نقوم أولا بإجراء تدريبات مسبقة من الناحية التي تم إشرافها ذاتيا على مجموعات بيانات النية التي تم جمعها، والتي تتعلم ضمنيا التمييز بين الكلام المماثلة الدلوية دون استخدام أي ملصقات.ثم نقوم بعد ذلك بإجراء اكتشاف عهد القليل من الرصاص مع التعلم البسيط المشروع، والذي يسحب صراحة النطق من نفس النية أقرب ويغطي الكلام عبر النوايا المختلفة أبعد.تظهر النتائج التجريبية أن أسلوبنا المقترح يحقق أداء حديثة على ثلاثة مجموعات بيانات للكشف عن النوايا الصعبة تحت 5 لقطة و 10 لقطة.
In this work, we focus on a more challenging few-shot intent detection scenario where many intents are fine-grained and semantically similar. We present a simple yet effective few-shot intent detection schema via contrastive pre-training and fine-tuning. Specifically, we first conduct self-supervised contrastive pre-training on collected intent datasets, which implicitly learns to discriminate semantically similar utterances without using any labels. We then perform few-shot intent detection together with supervised contrastive learning, which explicitly pulls utterances from the same intent closer and pushes utterances across different intents farther. Experimental results show that our proposed method achieves state-of-the-art performance on three challenging intent detection datasets under 5-shot and 10-shot settings.
References used
https://aclanthology.org/
As the labeling cost for different modules in task-oriented dialog (ToD) systems is expensive, a major challenge is to train different modules with the least amount of labeled data. Recently, large-scale pre-trained language models, have shown promis
This paper investigates the effectiveness of pre-training for few-shot intent classification. While existing paradigms commonly further pre-train language models such as BERT on a vast amount of unlabeled corpus, we find it highly effective and effic
We present VideoCLIP, a contrastive approach to pre-train a unified model for zero-shot video and text understanding, without using any labels on downstream tasks. VideoCLIP trains a transformer for video and text by contrasting temporally overlappin
In this paper we describe our submission to the multilingual Indic language translation wtask MultiIndicMT'' under the team name NICT-5''. This task involves translation from 10 Indic languages into English and vice-versa. The objective of the task w
Reading order detection is the cornerstone to understanding visually-rich documents (e.g., receipts and forms). Unfortunately, no existing work took advantage of advanced deep learning models because it is too laborious to annotate a large enough dat