Yanan Wu


2022

pdf
Distribution Calibration for Out-of-Domain Detection with Bayesian Approximation
Yanan Wu | Zhiyuan Zeng | Keqing He | Yutao Mou | Pei Wang | Weiran Xu
Proceedings of the 29th International Conference on Computational Linguistics

Out-of-Domain (OOD) detection is a key component in a task-oriented dialog system, which aims to identify whether a query falls outside the predefined supported intent set. Previous softmax-based detection algorithms are proved to be overconfident for OOD samples. In this paper, we analyze overconfident OOD comes from distribution uncertainty due to the mismatch between the training and test distributions, which makes the model can’t confidently make predictions thus probably causes abnormal softmax scores. We propose a Bayesian OOD detection framework to calibrate distribution uncertainty using Monte-Carlo Dropout. Our method is flexible and easily pluggable to existing softmax-based baselines and gains 33.33% OOD F1 improvements with increasing only 0.41% inference time compared to MSP. Further analyses show the effectiveness of Bayesian learning for OOD detection.

pdf
Generalized Intent Discovery: Learning from Open World Dialogue System
Yutao Mou | Keqing He | Yanan Wu | Pei Wang | Jingang Wang | Wei Wu | Yi Huang | Junlan Feng | Weiran Xu
Proceedings of the 29th International Conference on Computational Linguistics

Traditional intent classification models are based on a pre-defined intent set and only recognize limited in-domain (IND) intent classes. But users may input out-of-domain (OOD) queries in a practical dialogue system. Such OOD queries can provide directions for future improvement. In this paper, we define a new task, Generalized Intent Discovery (GID), which aims to extend an IND intent classifier to an open-world intent set including IND and OOD intents. We hope to simultaneously classify a set of labeled IND intent classes while discovering and recognizing new unlabeled OOD types incrementally. We construct three public datasets for different application scenarios and propose two kinds of frameworks, pipeline-based and end-to-end for future work. Further, we conduct exhaustive experiments and qualitative analysis to comprehend key challenges and provide new guidance for future GID research.

pdf
Revisit Overconfidence for OOD Detection: Reassigned Contrastive Learning with Adaptive Class-dependent Threshold
Yanan Wu | Keqing He | Yuanmeng Yan | QiXiang Gao | Zhiyuan Zeng | Fujia Zheng | Lulu Zhao | Huixing Jiang | Wei Wu | Weiran Xu
Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies

Detecting Out-of-Domain (OOD) or unknown intents from user queries is essential in a task-oriented dialog system. A key challenge of OOD detection is the overconfidence of neural models. In this paper, we comprehensively analyze overconfidence and classify it into two perspectives: over-confident OOD and in-domain (IND). Then according to intrinsic reasons, we respectively propose a novel reassigned contrastive learning (RCL) to discriminate IND intents for over-confident OOD and an adaptive class-dependent local threshold mechanism to separate similar IND and OOD intents for over-confident IND. Experiments and analyses show the effectiveness of our proposed method for both aspects of overconfidence issues.

pdf
Domain-Oriented Prefix-Tuning: Towards Efficient and Generalizable Fine-tuning for Zero-Shot Dialogue Summarization
Lulu Zhao | Fujia Zheng | Weihao Zeng | Keqing He | Weiran Xu | Huixing Jiang | Wei Wu | Yanan Wu
Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies

The most advanced abstractive dialogue summarizers lack generalization ability on new domains and the existing researches for domain adaptation in summarization generally rely on large-scale pre-trainings. To explore the lightweight fine-tuning methods for domain adaptation of dialogue summarization, in this paper, we propose an efficient and generalizable Domain-Oriented Prefix-tuning model, which utilizes a domain word initialized prefix module to alleviate domain entanglement and adopts discrete prompts to guide the model to focus on key contents of dialogues and enhance model generalization. We conduct zero-shot experiments and build domain adaptation benchmarks on two multi-domain dialogue summarization datasets, TODSum and QMSum. Adequate experiments and qualitative analysis prove the effectiveness of our methods.

pdf
Learning Discriminative Representations for Open Relation Extraction with Instance Ranking and Label Calibration
Shusen Wang | Bin Duan | Yanan Wu | Yajing Xu
Findings of the Association for Computational Linguistics: NAACL 2022

Open relation extraction is the task to extract relational facts without pre-defined relation types from open-domain corpora. However, since there are some hard or semi-hard instances sharing similar context and entity information but belonging to different underlying relation, current OpenRE methods always cluster them into the same relation type. In this paper, we propose a novel method based on Instance Ranking and Label Calibration strategies (IRLC) to learn discriminative representations for open relation extraction. Due to lacking the original instance label, we provide three surrogate strategies to generate the positive, hard negative, and semi-hard negative instances for the original instance. Instance ranking aims to refine the relational feature space by pushing the hard and semi-hard negative instances apart from the original instance with different margins and pulling the original instance and its positive instance together. To refine the cluster probability distributions of these instances, we introduce a label calibration strategy to model the constraint relationship between instances. Experimental results on two public datasets demonstrate that our proposed method can significantly outperform the previous state-of-the-art methods.

pdf
RCL: Relation Contrastive Learning for Zero-Shot Relation Extraction
Shusen Wang | Bosen Zhang | Yajing Xu | Yanan Wu | Bo Xiao
Findings of the Association for Computational Linguistics: NAACL 2022

Zero-shot relation extraction aims to identify novel relations which cannot be observed at the training stage. However, it still faces some challenges since the unseen relations of instances are similar or the input sentences have similar entities, the unseen relation representations from different categories tend to overlap and lead to errors. In this paper, we propose a novel Relation Contrastive Learning framework (RCL) to mitigate above two types of similar problems: Similar Relations and Similar Entities. By jointly optimizing a contrastive instance loss with a relation classification loss on seen relations, RCL can learn subtle difference between instances and achieve better separation between different relation categories in the representation space simultaneously. Especially in contrastive instance learning, the dropout noise as data augmentation is adopted to amplify the semantic difference between similar instances without breaking relation representation, so as to promote model to learn more effective representations. Experiments conducted on two well-known datasets show that RCL can significantly outperform previous state-of-the-art methods. Moreover, if the seen relations are insufficient, RCL can also obtain comparable results with the model trained on the full training set, showing the robustness of our approach.

pdf
Disentangling Confidence Score Distribution for Out-of-Domain Intent Detection with Energy-Based Learning
Yanan Wu | Zhiyuan Zeng | Keqing He | Yutao Mou | Pei Wang | Yuanmeng Yan | Weiran Xu
Proceedings of the Towards Semi-Supervised and Reinforced Task-Oriented Dialog Systems (SereTOD)

Detecting Out-of-Domain (OOD) or unknown intents from user queries is essential in a taskoriented dialog system. Traditional softmaxbased confidence scores are susceptible to the overconfidence issue. In this paper, we propose a simple but strong energy-based score function to detect OOD where the energy scores of OOD samples are higher than IND samples. Further, given a small set of labeled OOD samples, we introduce an energy-based margin objective for supervised OOD detection to explicitly distinguish OOD samples from INDs. Comprehensive experiments and analysis prove our method helps disentangle confidence score distributions of IND and OOD data.

pdf
Disentangled Knowledge Transfer for OOD Intent Discovery with Unified Contrastive Learning
Yutao Mou | Keqing He | Yanan Wu | Zhiyuan Zeng | Hong Xu | Huixing Jiang | Wei Wu | Weiran Xu
Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)

Discovering Out-of-Domain(OOD) intents is essential for developing new skills in a task-oriented dialogue system. The key challenge is how to transfer prior IND knowledge to OOD clustering. Different from existing work based on shared intent representation, we propose a novel disentangled knowledge transfer method via a unified multi-head contrastive learning framework. We aim to bridge the gap between IND pre-training and OOD clustering. Experiments and analysis on two benchmark datasets show the effectiveness of our method.

pdf
Watch the Neighbors: A Unified K-Nearest Neighbor Contrastive Learning Framework for OOD Intent Discovery
Yutao Mou | Keqing He | Pei Wang | Yanan Wu | Jingang Wang | Wei Wu | Weiran Xu
Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

Discovering out-of-domain (OOD) intent is important for developing new skills in task-oriented dialogue systems. The key challenges lie in how to transfer prior in-domain (IND) knowledge to OOD clustering, as well as jointly learn OOD representations and cluster assignments. Previous methods suffer from in-domain overfitting problem, and there is a natural gap between representation learning and clustering objectives. In this paper, we propose a unified K-nearest neighbor contrastive learning framework to discover OOD intents. Specifically, for IND pre-training stage, we propose a KCL objective to learn inter-class discriminative features, while maintaining intra-class diversity, which alleviates the in-domain overfitting problem. For OOD clustering stage, we propose a KCC method to form compact clusters by mining true hard negative samples, which bridges the gap between clustering and representation learning. Extensive experiments on three benchmark datasets show that our method achieves substantial improvements over the state-of-the-art methods.

pdf
UniNL: Aligning Representation Learning with Scoring Function for OOD Detection via Unified Neighborhood Learning
Yutao Mou | Pei Wang | Keqing He | Yanan Wu | Jingang Wang | Wei Wu | Weiran Xu
Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

Detecting out-of-domain (OOD) intents from user queries is essential for avoiding wrong operations in task-oriented dialogue systems. The key challenge is how to distinguish in-domain (IND) and OOD intents. Previous methods ignore the alignment between representation learning and scoring function, limiting the OOD detection performance. In this paper, we propose a unified neighborhood learning framework (UniNL) to detect OOD intents. Specifically, we design a KNCL objective for representation learning, and introduce a KNN-based scoring function for OOD detection. We aim to align representation learning with scoring function. Experiments and analysis on two benchmark datasets show the effectiveness of our method.

2021

pdf
Novel Slot Detection: A Benchmark for Discovering Unknown Slot Types in the Task-Oriented Dialogue System
Yanan Wu | Zhiyuan Zeng | Keqing He | Hong Xu | Yuanmeng Yan | Huixing Jiang | Weiran Xu
Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)

Existing slot filling models can only recognize pre-defined in-domain slot types from a limited slot set. In the practical application, a reliable dialogue system should know what it does not know. In this paper, we introduce a new task, Novel Slot Detection (NSD), in the task-oriented dialogue system. NSD aims to discover unknown or out-of-domain slot types to strengthen the capability of a dialogue system based on in-domain training data. Besides, we construct two public NSD datasets, propose several strong NSD baselines, and establish a benchmark for future work. Finally, we conduct exhaustive experiments and qualitative analysis to comprehend key challenges and provide new guidance for future directions.

pdf
Modeling Discriminative Representations for Out-of-Domain Detection with Supervised Contrastive Learning
Zhiyuan Zeng | Keqing He | Yuanmeng Yan | Zijun Liu | Yanan Wu | Hong Xu | Huixing Jiang | Weiran Xu
Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 2: Short Papers)

Detecting Out-of-Domain (OOD) or unknown intents from user queries is essential in a task-oriented dialog system. A key challenge of OOD detection is to learn discriminative semantic features. Traditional cross-entropy loss only focuses on whether a sample is correctly classified, and does not explicitly distinguish the margins between categories. In this paper, we propose a supervised contrastive learning objective to minimize intra-class variance by pulling together in-domain intents belonging to the same class and maximize inter-class variance by pushing apart samples from different classes. Besides, we employ an adversarial augmentation mechanism to obtain pseudo diverse views of a sample in the latent space. Experiments on two public datasets prove the effectiveness of our method capturing discriminative representations for OOD detection.

pdf
Dynamically Disentangling Social Bias from Task-Oriented Representations with Adversarial Attack
Liwen Wang | Yuanmeng Yan | Keqing He | Yanan Wu | Weiran Xu
Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies

Representation learning is widely used in NLP for a vast range of tasks. However, representations derived from text corpora often reflect social biases. This phenomenon is pervasive and consistent across different neural models, causing serious concern. Previous methods mostly rely on a pre-specified, user-provided direction or suffer from unstable training. In this paper, we propose an adversarial disentangled debiasing model to dynamically decouple social bias attributes from the intermediate representations trained on the main task. We aim to denoise bias information while training on the downstream task, rather than completely remove social bias and pursue static unbiased representations. Experiments show the effectiveness of our method, both on the effect of debiasing and the main task performance.