HIT-SCIR At MMNLU-22: Consistency Regularization For Multilingual Spoken Language Understanding

Abstract

Multilingual spoken language understanding (SLU) consists of two sub-tasks, namely intent detection and slot filling. To improve the performance of these two sub-tasks, we propose to use consistency regularization based on a hybrid data augmentation strategy. The consistency regularization enforces the predicted distributions for an example and its semantically equivalent augmentation to be consistent. We conduct experiments on the MASSIVE dataset under both full-dataset and zero-shot settings. Experimental results demonstrate that our proposed method improves the performance on both intent detection and slot filling tasks. Our system\footnote\{The code will be available at https://github.com/bozheng-hit/MMNLU-22-HIT-SCIR.\} ranked 1st in the MMNLU-22 competition under the full-dataset setting.

Stats

• citations1
• S2 citations—
• github stars0
• HF likes0
• heat score2.26
• arxiv keyzheng2023hit

Code

• bozheng-hit/MMNLU-22-HIT-SCIR—

Related papers

• Modality Confidence Aware Training For Robust End-to-end Spoken Language Understanding (2023)2.26
• ML-LMCL: Mutual Learning And Large-margin Contrastive Learning For Improving ASR Robustness In Spoken Language Understanding (2023)0.00
• Unislu: Unified Spoken Language Understanding From Heterogeneous Cross-task Datasets (2025)0.00
• Contrastive Learning For Improving ASR Robustness In Spoken Language Understanding (2022)6.34
• Cross-lingual Spoken Language Understanding With Regularized Representation Alignment (2020)6.77
• Bridging The Gap Between Clean Data Training And Real-world Inference For Spoken Language Understanding (2021)0.00
• Data Augmentation For Spoken Language Understanding Via Pretrained Language Models (2020)0.00
• Label-aware Multi-level Contrastive Learning For Cross-lingual Spoken Language Understanding (2022)6.34