# A Two-Stage Prediction-Aware Contrastive Learning Framework for Multi-Intent NLU

arXiv: https://arxiv.org/abs/2405.02925 · HTML: https://mlsanity.com/paper/2405.02925/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 33 references that are in the map.
- First posted on arXiv 5 May 2024.

## Builds on (most cited first, 20 of 20)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [SimCSE: Simple Contrastive Learning of Sentence Embeddings](https://mlsanity.com/paper/2104.08821/) (2021)
- [Supervised Contrastive Learning for Pre-trained Language Model Fine-tuning](https://mlsanity.com/paper/2011.01403/) (2020)
- [BERT for Joint Intent Classification and Slot Filling](https://mlsanity.com/paper/1902.10909/) (2019)
- [A Stack-Propagation Framework with Token-Level Intent Detection for Spoken Language Understanding](https://mlsanity.com/paper/1909.02188/) (2019)
- [A Novel Bi-directional Interrelated Model for Joint Intent Detection and Slot Filling](https://mlsanity.com/paper/1907.00390/) (2019)
- [Joint Slot Filling and Intent Detection via Capsule Neural Networks](https://mlsanity.com/paper/1812.09471/) (2018)
- [SlotRefine: A Fast Non-Autoregressive Model for Joint Intent Detection and Slot Filling](https://mlsanity.com/paper/2010.02693/) (2020)
- [GL-GIN: Fast and Accurate Non-Autoregressive Model for Joint Multiple Intent Detection and Slot Filling](https://mlsanity.com/paper/2106.01925/) (2021)
- [Meta-Curriculum Learning for Domain Adaptation in Neural Machine Translation](https://mlsanity.com/paper/2103.02262/) (2021)
- [Few-shot Learning for Multi-label Intent Detection](https://mlsanity.com/paper/2010.05256/) (2020)
- [Joint Multiple Intent Detection and Slot Filling via Self-distillation](https://mlsanity.com/paper/2108.08042/) (2021)
- [SLIM: Explicit Slot-Intent Mapping with BERT for Joint Multi-Intent Detection and Slot Filling](https://mlsanity.com/paper/2108.11711/) (2021)
- [An Explicit-Joint and Supervised-Contrastive Learning Framework for Few-Shot Intent Classification and Slot Filling](https://mlsanity.com/paper/2110.13691/) (2021)
- [Joint Multiple Intent Detection and Slot Filling with Supervised Contrastive Learning and Self-Distillation](https://mlsanity.com/paper/2308.14654/) (2023)
- [Learning to Bridge Metric Spaces: Few-shot Joint Learning of Intent Detection and Slot Filling](https://mlsanity.com/paper/2106.07343/) (2021)
- [An Effective Non-Autoregressive Model for Spoken Language Understanding](https://mlsanity.com/paper/2108.07005/) (2021)
- [QAID: Question Answering Inspired Few-shot Intent Detection](https://mlsanity.com/paper/2303.01593/) (2023)

## Built on by (most cited first, 2 of 2)

- [MIDAS: Multi-level Intent, Domain, And Slot Knowledge Distillation for Multi-turn NLU](https://mlsanity.com/paper/2408.08144/) (2024)
- [Before the Arrest: Benchmarking LLMs on Criminal Profiling from Incomplete Evidence](https://mlsanity.com/paper/2609.19965/) (2026)

## Abstract (arXiv)

Multi-intent natural language understanding (NLU) presents a formidable challenge due to the model confusion arising from multiple intents within a single utterance. While previous works train the model contrastively to increase the margin between different multi-intent labels, they are less suited to the nuances of multi-intent NLU. They ignore the rich information between the shared intents, which is beneficial to constructing a better embedding space, especially in low-data scenarios. We introduce a two-stage Prediction-Aware Contrastive Learning (PACL) framework for multi-intent NLU to harness this valuable knowledge. Our approach capitalizes on shared intent information by integrating word-level pre-training and prediction-aware contrastive fine-tuning. We construct a pre-training dataset using a word-level data augmentation strategy. Subsequently, our framework dynamically assigns roles to instances during contrastive fine-tuning while introducing a prediction-aware contrastive loss to maximize the impact of contrastive learning. We present experimental results and empirical analysis conducted on three widely used datasets, demonstrating that our method surpasses the performance of three prominent baselines on both low-data and full-data scenarios.