Yan Liu


2023

pdf bib
DSRM: Boost Textual Adversarial Training with Distribution Shift Risk Minimization
SongYang Gao | Shihan Dou | Yan Liu | Xiao Wang | Qi Zhang | Zhongyu Wei | Jin Ma | Ying Shan
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

Adversarial training is one of the best-performing methods in improving the robustness of deep language models. However, robust models come at the cost of high time consumption, as they require multi-step gradient ascents or word substitutions to obtain adversarial samples. In addition, these generated samples are deficient in grammatical quality and semantic consistency, which impairs the effectiveness of adversarial training. To address these problems, we introduce a novel, effective procedure for instead adversarial training with only clean data. Our procedure, distribution shift risk minimization (DSRM), estimates the adversarial loss by perturbing the input data’s probability distribution rather than their embeddings. This formulation results in a robust model that minimizes the expected global loss under adversarial attacks. Our approach requires zero adversarial samples for training and reduces time consumption by up to 70% compared to current best-performing adversarial training methods. Experiments demonstrate that DSRM considerably improves BERT’s resistance to textual adversarial attacks and achieves state-of-the-art robust accuracy on various benchmarks.

pdf bib
Uncovering and Categorizing Social Biases in Text-to-SQL
Yan Liu | Yan Gao | Zhe Su | Xiaokang Chen | Elliott Ash | Jian-Guang Lou
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

Large pre-trained language models are acknowledged to carry social bias towards different demographics, which can further amplify existing stereotypes in our society and cause even more harm. Text-to-SQL is an important task, models of which are mainly adopted by administrative industries, where unfair decisions may lead to catastrophic consequences. However, existing Text-to-SQL models are trained on clean, neutral datasets, such as Spider and WikiSQL. This, to some extent, cover up social bias in models under ideal conditions, which nevertheless may emerge in real application scenarios. In this work, we aim to uncover and mitigate social bias in Text-to-SQL models. We summarize the categories of social bias that may occur in structural data for Text-to-SQL models. We build test benchmarks and reveal that models with similar task accuracy can contain social bias at very different rates. We show how to take advantage of our methodology to assess and mitigate social bias in the downstream Text-to-SQL task.

2022

pdf bib
MPII: Multi-Level Mutual Promotion for Inference and Interpretation
Yan Liu | Sanyuan Chen | Yazheng Yang | Qi Dai
Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

In order to better understand the rationale behind model behavior, recent works have exploited providing interpretation to support the inference prediction. However, existing methods tend to provide human-unfriendly interpretation, and are prone to sub-optimal performance due to one-side promotion, i.e. either inference promotion with interpretation or vice versa. In this paper, we propose a multi-level Mutual Promotion mechanism for self-evolved Inference and sentence-level Interpretation (MPII). Specifically, from the model-level, we propose a Step-wise Integration Mechanism to jointly perform and deeply integrate inference and interpretation in an autoregressive manner. From the optimization-level, we propose an Adversarial Fidelity Regularization to improve the fidelity between inference and interpretation with the Adversarial Mutual Information training strategy. Extensive experiments on NLI and CQA tasks reveal that the proposed MPII approach can significantly outperform baseline models for both the inference performance and the interpretation quality.

2019

pdf bib
GumDrop at the DISRPT2019 Shared Task: A Model Stacking Approach to Discourse Unit Segmentation and Connective Detection
Yue Yu | Yilun Zhu | Yang Liu | Yan Liu | Siyao Peng | Mackenzie Gong | Amir Zeldes
Proceedings of the Workshop on Discourse Relation Parsing and Treebanking 2019

In this paper we present GumDrop, Georgetown University’s entry at the DISRPT 2019 Shared Task on automatic discourse unit segmentation and connective detection. Our approach relies on model stacking, creating a heterogeneous ensemble of classifiers, which feed into a metalearner for each final task. The system encompasses three trainable component stacks: one for sentence splitting, one for discourse unit segmentation and one for connective detection. The flexibility of each ensemble allows the system to generalize well to datasets of different sizes and with varying levels of homogeneity.

2016

pdf bib
Universal dependencies for Uyghur
Marhaba Eli | Weinila Mushajiang | Tuergen Yibulayin | Kahaerjiang Abiderexiti | Yan Liu
Proceedings of the Third International Workshop on Worldwide Language Service Infrastructure and Second Workshop on Open Infrastructures and Analysis Frameworks for Human Language Technologies (WLSI/OIAF4HLT2016)

The Universal Dependencies (UD) Project seeks to build a cross-lingual studies of treebanks, linguistic structures and parsing. Its goal is to create a set of multilingual harmonized treebanks that are designed according to a universal annotation scheme. In this paper, we report on the conversion of the Uyghur dependency treebank to a UD version of the treebank which we term the Uyghur Universal Dependency Treebank (UyDT). We present the mapping of the Uyghur dependency treebank’s labelling scheme to the UD scheme, along with a clear description of the structural changes required in this conversion.

2010

pdf bib
Exploring Deep Belief Network for Chinese Relation Extraction
Yu Chen | Wenjie Li | Yan Liu | Dequan Zheng | Tiejun Zhao
CIPS-SIGHAN Joint Conference on Chinese Language Processing