凶兆
发表于 2025-3-23 13:27:02
http://reply.papertrans.cn/48/4704/470333/470333_11.png
其他
发表于 2025-3-23 17:41:31
http://reply.papertrans.cn/48/4704/470333/470333_12.png
FLIRT
发表于 2025-3-23 18:41:14
http://reply.papertrans.cn/48/4704/470333/470333_13.png
Accrue
发表于 2025-3-24 00:23:29
Dialogue System quality and providing personalized services. Existing research on new intent discovery is divided into three schools based on unknown intent detection, non-clustering, and semi-clustering, but accurately identifying and understanding new types of user intents is still challenging. Efficiently ident
轨道
发表于 2025-3-24 04:57:11
Intent Recognitionof detecting new user intents that do not appear in the pre-defined intent set is called “unknown intent detection.” After successfully separating unknown intents from known intents, more attention is given to what new intents have been discovered. The process of classifying unknown intents into new
CRASS
发表于 2025-3-24 07:44:35
http://reply.papertrans.cn/48/4704/470333/470333_16.png
描绘
发表于 2025-3-24 14:04:26
http://reply.papertrans.cn/48/4704/470333/470333_17.png
Cocker
发表于 2025-3-24 17:13:29
Unknown Intent Detection Method Based on Model Post-Processingks are fed into a traditional novelty detection algorithm to detect unknown intents from different perspectives. Finally, the methods mentioned above are combined to facilitate joint prediction. The proposed method classifies examples that differ from known intents as unknown and does not require an
邪恶的你
发表于 2025-3-24 22:13:03
Discovering New Intents with Deep Aligned Clusteringer of intent categories is predicted by eliminating low-confidence intent-wise clusters. Extensive experiments on two benchmark datasets show that the method presented is more robust and achieves substantial improvements over the state-of-the-art methods.
legitimate
发表于 2025-3-25 02:04:15
http://reply.papertrans.cn/48/4704/470333/470333_20.png