Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2404.08977 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911844708384768 |
|---|---|
| author | Zhang, Shun Yan, Chaoran Yang, Jian Ren, Changyu Bai, Jiaqi Li, Tongliang Li, Zhoujun |
| author_facet | Zhang, Shun Yan, Chaoran Yang, Jian Ren, Changyu Bai, Jiaqi Li, Tongliang Li, Zhoujun |
| contents | New Intent Discovery (NID) strives to identify known and reasonably deduce novel intent groups in the open-world scenario. But current methods face issues with inaccurate pseudo-labels and poor representation learning, creating a negative feedback loop that degrades overall model performance, including accuracy and the adjusted rand index. To address the aforementioned challenges, we propose a Robust New Intent Discovery (RoNID) framework optimized by an EM-style method, which focuses on constructing reliable pseudo-labels and obtaining cluster-friendly discriminative representations. RoNID comprises two main modules: reliable pseudo-label generation module and cluster-friendly representation learning module. Specifically, the pseudo-label generation module assigns reliable synthetic labels by solving an optimal transport problem in the E-step, which effectively provides high-quality supervised signals for the input of the cluster-friendly representation learning module. To learn cluster-friendly representation with strong intra-cluster compactness and large inter-cluster separation, the representation learning module combines intra-cluster and inter-cluster contrastive learning in the M-step to feed more discriminative features into the generation module. RoNID can be performed iteratively to ultimately yield a robust model with reliable pseudo-labels and cluster-friendly representations. Experimental results on multiple benchmarks demonstrate our method brings substantial improvements over previous state-of-the-art methods by a large margin of +1~+4 points. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_08977 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | RoNID: New Intent Discovery with Generated-Reliable Labels and Cluster-friendly Representations Zhang, Shun Yan, Chaoran Yang, Jian Ren, Changyu Bai, Jiaqi Li, Tongliang Li, Zhoujun Computation and Language Machine Learning New Intent Discovery (NID) strives to identify known and reasonably deduce novel intent groups in the open-world scenario. But current methods face issues with inaccurate pseudo-labels and poor representation learning, creating a negative feedback loop that degrades overall model performance, including accuracy and the adjusted rand index. To address the aforementioned challenges, we propose a Robust New Intent Discovery (RoNID) framework optimized by an EM-style method, which focuses on constructing reliable pseudo-labels and obtaining cluster-friendly discriminative representations. RoNID comprises two main modules: reliable pseudo-label generation module and cluster-friendly representation learning module. Specifically, the pseudo-label generation module assigns reliable synthetic labels by solving an optimal transport problem in the E-step, which effectively provides high-quality supervised signals for the input of the cluster-friendly representation learning module. To learn cluster-friendly representation with strong intra-cluster compactness and large inter-cluster separation, the representation learning module combines intra-cluster and inter-cluster contrastive learning in the M-step to feed more discriminative features into the generation module. RoNID can be performed iteratively to ultimately yield a robust model with reliable pseudo-labels and cluster-friendly representations. Experimental results on multiple benchmarks demonstrate our method brings substantial improvements over previous state-of-the-art methods by a large margin of +1~+4 points. |
| title | RoNID: New Intent Discovery with Generated-Reliable Labels and Cluster-friendly Representations |
| topic | Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2404.08977 |