Search Results for author: Minlie Huang

Found 198 papers, 123 papers with code

Persona-Guided Planning for Controlling the Protagonist’s Persona in Story Generation

1 code implementation NAACL 2022 Zhexin Zhang, Jiaxin Wen, Jian Guan, Minlie Huang

In this paper, we aim to control the protagonist’s persona in story generation, i. e., generating a story from a leading context and a persona description, where the protagonist should exhibit the specified personality through a coherent event sequence.

Sentence Story Generation

Turn-Level User Satisfaction Estimation in E-commerce Customer Service

no code implementations ACL (ECNLP) 2021 Runze Liang, Ryuichi Takanobu, Feng-Lin Li, Ji Zhang, Haiqing Chen, Minlie Huang

To this end, we formalize the turn-level satisfaction estimation as a reinforcement learning problem, in which the model can be optimized with only session-level satisfaction labels.

When does Further Pre-training MLM Help? An Empirical Study on Task-Oriented Dialog Pre-training

1 code implementation EMNLP (insights) 2021 Qi Zhu, Yuxian Gu, Lingxiao Luo, Bing Li, Cheng Li, Wei Peng, Minlie Huang, Xiaoyan Zhu

Further pre-training language models on in-domain data (domain-adaptive pre-training, DAPT) or task-relevant data (task-adaptive pre-training, TAPT) before fine-tuning has been shown to improve downstream tasks’ performances.

Learning Task Decomposition to Assist Humans in Competitive Programming

no code implementations7 Jun 2024 Jiaxin Wen, Ruiqi Zhong, Pei Ke, Zhihong Shao, Hongning Wang, Minlie Huang

When using language models (LMs) to solve complex problems, humans might struggle to understand the LM-generated solutions and repair the flawed ones.

Weak-to-Strong Extrapolation Expedites Alignment

1 code implementation25 Apr 2024 Chujie Zheng, Ziqi Wang, Heng Ji, Minlie Huang, Nanyun Peng

Through experiments with twelve open-source LLMs on HuggingFace, we demonstrate that ExPO consistently improves off-the-shelf DPO/RLHF models, as evaluated on the mainstream LLM benchmarks AlpacaEval 2. 0 and MT-Bench.

360°REA: Towards A Reusable Experience Accumulation with 360° Assessment for Multi-Agent System

no code implementations8 Apr 2024 Shen Gao, Hao Li, Zhengliang Shi, Chengrui Huang, Quan Tu, Zhiliang Tian, Minlie Huang, Shuo Shang

The framework employs a novel 360{\deg} performance assessment method for multi-perspective performance evaluation with fine-grained assessment.

Language Modelling Large Language Model

ChatGLM-RLHF: Practices of Aligning Large Language Models with Human Feedback

no code implementations1 Apr 2024 Zhenyu Hou, Yilin Niu, Zhengxiao Du, Xiaohan Zhang, Xiao Liu, Aohan Zeng, Qinkai Zheng, Minlie Huang, Hongning Wang, Jie Tang, Yuxiao Dong

The work presents our practices of aligning LLMs with human preferences, offering insights into the challenges and solutions in RLHF implementations.

Towards Optimal Learning of Language Models

no code implementations27 Feb 2024 Yuxian Gu, Li Dong, Yaru Hao, Qingxiu Dong, Minlie Huang, Furu Wei

This work studies the general principles of improving the learning of language models (LMs), which aims at reducing the necessary training steps for achieving superior performance.

Data Compression Language Modelling

ShieldLM: Empowering LLMs as Aligned, Customizable and Explainable Safety Detectors

1 code implementation26 Feb 2024 Zhexin Zhang, Yida Lu, Jingyuan Ma, Di Zhang, Rui Li, Pei Ke, Hao Sun, Lei Sha, Zhifang Sui, Hongning Wang, Minlie Huang

The safety of Large Language Models (LLMs) has gained increasing attention in recent years, but there still lacks a comprehensive approach for detecting safety issues within LLMs' responses in an aligned, customizable and explainable manner.

LLM-based Privacy Data Augmentation Guided by Knowledge Distillation with a Distribution Tutor for Medical Text Classification

no code implementations26 Feb 2024 Yiping Song, Juhua Zhang, Zhiliang Tian, Yuxin Yang, Minlie Huang, Dongsheng Li

As sufficient data are not always publically accessible for model training, researchers exploit limited data with advanced learning algorithms or expand the dataset via data augmentation (DA).

Data Augmentation Knowledge Distillation +2

ASETF: A Novel Method for Jailbreak Attack on LLMs through Translate Suffix Embeddings

no code implementations25 Feb 2024 Hao Wang, Hao Li, Minlie Huang, Lei Sha

In addition, our approach can be generalized into a broader method for generating transferable adversarial suffixes that can successfully attack multiple LLMs, even black-box LLMs, such as ChatGPT and Gemini.

Language Modelling Large Language Model

AMOR: A Recipe for Building Adaptable Modular Knowledge Agents Through Process Feedback

no code implementations2 Feb 2024 Jian Guan, Wei Wu, Zujie Wen, Peng Xu, Hongning Wang, Minlie Huang

We present AMOR, an agent framework based on open-source LLMs, which reasons with external knowledge bases and adapts to specific domains through human supervision to the reasoning process.

Towards Efficient Exact Optimization of Language Model Alignment

2 code implementations1 Feb 2024 Haozhe Ji, Cheng Lu, Yilin Niu, Pei Ke, Hongning Wang, Jun Zhu, Jie Tang, Minlie Huang

This leads to the same mode-seeking solution, while enables efficient optimization by circumventing the complexities of RL.

Language Modelling Reinforcement Learning (RL)

On Prompt-Driven Safeguarding for Large Language Models

1 code implementation31 Jan 2024 Chujie Zheng, Fan Yin, Hao Zhou, Fandong Meng, Jie zhou, Kai-Wei Chang, Minlie Huang, Nanyun Peng

In this work, we investigate how LLMs' behavior (i. e., complying with or refusing user queries) is affected by safety prompts from the perspective of model representation.

CritiqueLLM: Scaling LLM-as-Critic for Effective and Explainable Evaluation of Large Language Model Generation

2 code implementations30 Nov 2023 Pei Ke, Bosi Wen, Zhuoer Feng, Xiao Liu, Xuanyu Lei, Jiale Cheng, Shengyuan Wang, Aohan Zeng, Yuxiao Dong, Hongning Wang, Jie Tang, Minlie Huang

Since the natural language processing (NLP) community started to make large language models (LLMs), such as GPT-4, act as a critic to evaluate the quality of generated texts, most of them only train a critique generation model of a specific scale on specific datasets.

Language Modelling Large Language Model

Unveiling the Implicit Toxicity in Large Language Models

1 code implementation29 Nov 2023 Jiaxin Wen, Pei Ke, Hao Sun, Zhexin Zhang, Chengfei Li, Jinfeng Bai, Minlie Huang

While recent studies primarily focus on probing toxic outputs that can be easily detected with existing toxicity classifiers, we show that LLMs can generate diverse implicit toxic outputs that are exceptionally difficult to detect via simply zero-shot prompting.

Language Modelling Reinforcement Learning (RL)

Defending Large Language Models Against Jailbreaking Attacks Through Goal Prioritization

1 code implementation15 Nov 2023 Zhexin Zhang, Junxiao Yang, Pei Ke, Fei Mi, Hongning Wang, Minlie Huang

While significant attention has been dedicated to exploiting weaknesses in LLMs through jailbreaking attacks, there remains a paucity of effort in defending against these attacks.

Instruction Following

Black-Box Prompt Optimization: Aligning Large Language Models without Model Training

1 code implementation7 Nov 2023 Jiale Cheng, Xiao Liu, Kehan Zheng, Pei Ke, Hongning Wang, Yuxiao Dong, Jie Tang, Minlie Huang

However, these models are often not well aligned with human intents, which calls for additional treatments on them, that is, the alignment problem.

Language Models Hallucinate, but May Excel at Fact Verification

1 code implementation23 Oct 2023 Jian Guan, Jesse Dodge, David Wadden, Minlie Huang, Hao Peng

Recent progress in natural language processing (NLP) owes much to remarkable advances in large language models (LLMs).

Fact Verification Hallucination

Task-Adaptive Tokenization: Enhancing Long-Form Text Generation Efficacy in Mental Health and Beyond

no code implementations9 Oct 2023 Siyang Liu, Naihao Deng, Sahand Sabour, Yilin Jia, Minlie Huang, Rada Mihalcea

We propose task-adaptive tokenization as a way to adapt the generation pipeline to the specifics of a downstream task and enhance long-form generation in mental health.

Question Answering Text Generation

Language Model Decoding as Direct Metrics Optimization

no code implementations2 Oct 2023 Haozhe Ji, Pei Ke, Hongning Wang, Minlie Huang

And most importantly, we prove that this induced distribution is guaranteed to improve the perplexity on human texts, which suggests a better approximation to the underlying distribution of human texts.

Language Modelling

ToRA: A Tool-Integrated Reasoning Agent for Mathematical Problem Solving

1 code implementation29 Sep 2023 Zhibin Gou, Zhihong Shao, Yeyun Gong, Yelong Shen, Yujiu Yang, Minlie Huang, Nan Duan, Weizhu Chen

Large language models have made significant progress in various language tasks, yet they still struggle with complex mathematics.

Ranked #11 on Math Word Problem Solving on MATH (using extra training data)

Arithmetic Reasoning Computational Efficiency +3

Large Language Models Are Not Robust Multiple Choice Selectors

1 code implementation7 Sep 2023 Chujie Zheng, Hao Zhou, Fandong Meng, Jie zhou, Minlie Huang

This work shows that modern LLMs are vulnerable to option position changes in MCQs due to their inherent "selection bias", namely, they prefer to select specific option IDs as answers (like "Option A").

Computational Efficiency Multiple-choice +1

AgentBench: Evaluating LLMs as Agents

1 code implementation7 Aug 2023 Xiao Liu, Hao Yu, Hanchen Zhang, Yifan Xu, Xuanyu Lei, Hanyu Lai, Yu Gu, Hangliang Ding, Kaiwen Men, Kejuan Yang, Shudan Zhang, Xiang Deng, Aohan Zeng, Zhengxiao Du, Chenhui Zhang, Sheng Shen, Tianjun Zhang, Yu Su, Huan Sun, Minlie Huang, Yuxiao Dong, Jie Tang

We present AgentBench, a multi-dimensional evolving benchmark that currently consists of 8 distinct environments to assess LLM-as-Agent's reasoning and decision-making abilities in a multi-turn open-ended generation setting.

Decision Making Instruction Following

Facilitating Multi-turn Emotional Support Conversation with Positive Emotion Elicitation: A Reinforcement Learning Approach

1 code implementation16 Jul 2023 Jinfeng Zhou, Zhuang Chen, Bo wang, Minlie Huang

Experiments verify the superiority of Supporter in achieving positive emotion elicitation during responding while maintaining conversational goals including coherence.

DecompEval: Evaluating Generated Texts as Unsupervised Decomposed Question Answering

1 code implementation13 Jul 2023 Pei Ke, Fei Huang, Fei Mi, Yasheng Wang, Qun Liu, Xiaoyan Zhu, Minlie Huang

Existing evaluation metrics for natural language generation (NLG) tasks face the challenges on generalization ability and interpretability.

Dialogue Generation nlg evaluation +3

Ethicist: Targeted Training Data Extraction Through Loss Smoothed Soft Prompting and Calibrated Confidence Estimation

1 code implementation10 Jul 2023 Zhexin Zhang, Jiaxin Wen, Minlie Huang

In this paper, we propose a method named Ethicist for targeted training data extraction through loss smoothed soft prompting and calibrated confidence estimation, investigating how to recover the suffix in the training data when given a prefix.

Memorization

Mitigating the Learning Bias towards Repetition by Self-Contrastive Training for Open-Ended Generation

1 code implementation4 Jul 2023 Jian Guan, Minlie Huang

Despite the huge progress in myriad generation tasks, pretrained language models (LMs) such as GPT2 still tend to generate repetitive texts with maximization-based decoding algorithms for open-ended generation.

Attribute Sentence

Click: Controllable Text Generation with Sequence Likelihood Contrastive Learning

1 code implementation6 Jun 2023 Chujie Zheng, Pei Ke, Zheng Zhang, Minlie Huang

It has always been an important yet challenging problem to control language models to avoid generating texts with undesirable attributes, such as toxic language and unnatural repetition.

Contrastive Learning Text Generation

Uncertainty in Natural Language Processing: Sources, Quantification, and Applications

no code implementations5 Jun 2023 Mengting Hu, Zhen Zhang, Shiwan Zhao, Minlie Huang, Bingzhe Wu

Therefore, in this survey, we provide a comprehensive review of uncertainty-relevant works in the NLP field.

Uncertainty Quantification

E-NER: Evidential Deep Learning for Trustworthy Named Entity Recognition

1 code implementation29 May 2023 Zhen Zhang, Mengting Hu, Shiwan Zhao, Minlie Huang, Haotian Wang, Lemao Liu, Zhirui Zhang, Zhe Liu, Bingzhe Wu

Most named entity recognition (NER) systems focus on improving model performance, ignoring the need to quantify model uncertainty, which is critical to the reliability of NER systems in open environments.

named-entity-recognition Named Entity Recognition +1

Enhancing Retrieval-Augmented Large Language Models with Iterative Retrieval-Generation Synergy

no code implementations24 May 2023 Zhihong Shao, Yeyun Gong, Yelong Shen, Minlie Huang, Nan Duan, Weizhu Chen

In this paper, we show that strong performance can be achieved by a method we call Iter-RetGen, which synergizes retrieval and generation in an iterative manner.

Fact Verification Multi-hop Question Answering +2

Pre-Training to Learn in Context

1 code implementation16 May 2023 Yuxian Gu, Li Dong, Furu Wei, Minlie Huang

In-context learning, where pre-trained language models learn to perform tasks from task examples and instructions in their contexts, has attracted much attention in the NLP community.

In-Context Learning Language Modelling +3

COKE: A Cognitive Knowledge Graph for Machine Theory of Mind

1 code implementation9 May 2023 Jincenzi Wu, Zhuang Chen, Jiawen Deng, Sahand Sabour, Helen Meng, Minlie Huang

Theory of mind (ToM) refers to humans' ability to understand and infer the desires, beliefs, and intentions of others.

Re$^3$Dial: Retrieve, Reorganize and Rescale Dialogue Corpus for Long-Turn Open-Domain Dialogue Pre-training

1 code implementation4 May 2023 Jiaxin Wen, Hao Zhou, Jian Guan, Minlie Huang

However, the pre-trained dialogue model's ability to utilize long-range context is limited due to the scarcity of long-turn dialogue sessions.

Directed Acyclic Transformer Pre-training for High-quality Non-autoregressive Text Generation

1 code implementation24 Apr 2023 Fei Huang, Pei Ke, Minlie Huang

Non-AutoRegressive (NAR) text generation models have drawn much attention because of their significantly faster decoding speed and good generation quality in machine translation.

Machine Translation Text Generation

Safety Assessment of Chinese Large Language Models

2 code implementations20 Apr 2023 Hao Sun, Zhexin Zhang, Jiawen Deng, Jiale Cheng, Minlie Huang

To further promote the safe deployment of LLMs, we develop a Chinese LLM safety assessment benchmark.

Tailoring Language Generation Models under Total Variation Distance

1 code implementation26 Feb 2023 Haozhe Ji, Pei Ke, Zhipeng Hu, Rongsheng Zhang, Minlie Huang

The standard paradigm of neural language generation adopts maximum likelihood estimation (MLE) as the optimizing method.

Text Generation

Towards Safer Generative Language Models: A Survey on Safety Risks, Evaluations, and Improvements

no code implementations18 Feb 2023 Jiawen Deng, Jiale Cheng, Hao Sun, Zhexin Zhang, Minlie Huang

This survey presents a framework for safety research pertaining to large models, delineating the landscape of safety risks as well as safety evaluation and improvement methods.

Adversarial Attack Ethics

Synthetic Prompting: Generating Chain-of-Thought Demonstrations for Large Language Models

no code implementations1 Feb 2023 Zhihong Shao, Yeyun Gong, Yelong Shen, Minlie Huang, Nan Duan, Weizhu Chen

However, the quality of the prompts depends on the demonstrations given to the models, and creating many of them by hand is costly.

PAL: Persona-Augmented Emotional Support Conversation Generation

1 code implementation19 Dec 2022 Jiale Cheng, Sahand Sabour, Hao Sun, Zhuang Chen, Minlie Huang

As previous studies have demonstrated that seekers' persona is an important factor for effective support, we investigate whether there are benefits to modeling such information in dialogue models for support.

Constructing Highly Inductive Contexts for Dialogue Safety through Controllable Reverse Generation

1 code implementation4 Dec 2022 Zhexin Zhang, Jiale Cheng, Hao Sun, Jiawen Deng, Fei Mi, Yasheng Wang, Lifeng Shang, Minlie Huang

In order to detect such toxic generations, existing methods rely on templates, real-world data extraction, crowdsourcing workers, or automatic generation to construct adversarial contexts that are likely to induce toxic generations.

Response Generation

AutoCAD: Automatically Generating Counterfactuals for Mitigating Shortcut Learning

1 code implementation29 Nov 2022 Jiaxin Wen, Yeshuang Zhu, Jinchao Zhang, Jie zhou, Minlie Huang

Recent studies have shown the impressive efficacy of counterfactually augmented data (CAD) for reducing NLU models' reliance on spurious features and improving their generalizability.

Chaining Simultaneous Thoughts for Numerical Reasoning

no code implementations29 Nov 2022 Zhihong Shao, Fei Huang, Minlie Huang

Given that rich information is hidden behind ubiquitous numbers in text, numerical reasoning over text should be an essential skill of AI systems.

Aligning Recommendation and Conversation via Dual Imitation

no code implementations5 Nov 2022 Jinfeng Zhou, Bo wang, Minlie Huang, Dongming Zhao, Kun Huang, Ruifang He, Yuexian Hou

Human conversations of recommendation naturally involve the shift of interests which can align the recommendation actions and conversation process to make accurate recommendations with rich explanations.

Recommendation Systems

Learning Instructions with Unlabeled Data for Zero-Shot Cross-Task Generalization

1 code implementation17 Oct 2022 Yuxian Gu, Pei Ke, Xiaoyan Zhu, Minlie Huang

Recently, instruction tuning (IT), which fine-tunes a pre-trained language model on a massive collection of tasks described via human-craft instructions, has been shown effective in instruction learning for unseen tasks.

Language Modelling

A Benchmark for Understanding and Generating Dialogue between Characters in Stories

no code implementations18 Sep 2022 Jianzhu Yao, Ziqi Liu, Jian Guan, Minlie Huang

We build a new dataset DialStory, which consists of 105k Chinese stories with a large amount of dialogue weaved into the plots to support the evaluation.

Dialogue Generation Speaker Recognition

StoryTrans: Non-Parallel Story Author-Style Transfer with Discourse Representations and Content Enhancing

1 code implementation29 Aug 2022 Xuekai Zhu, Jian Guan, Minlie Huang, Juan Liu

Moreover, to enhance content preservation, we design a mask-and-fill framework to explicitly fuse style-specific keywords of source texts into generation.

Sentence Style Transfer +1

CASE: Aligning Coarse-to-Fine Cognition and Affection for Empathetic Response Generation

1 code implementation18 Aug 2022 Jinfeng Zhou, Chujie Zheng, Bo wang, Zheng Zhang, Minlie Huang

Empathetic conversation is psychologically supposed to be the result of conscious alignment and interaction between the cognition and affection of empathy.

Dialogue Generation Empathetic Response Generation +1

Generating Coherent Narratives by Learning Dynamic and Discrete Entity States with a Contrastive Framework

1 code implementation8 Aug 2022 Jian Guan, Zhenyu Yang, Rongsheng Zhang, Zhipeng Hu, Minlie Huang

Despite advances in generating fluent texts, existing pretraining models tend to attach incoherent event sequences to involved entities when generating narratives such as stories and news.

Decoder Sentence

On the Learning of Non-Autoregressive Transformers

no code implementations13 Jun 2022 Fei Huang, Tianhua Tao, Hao Zhou, Lei LI, Minlie Huang

Non-autoregressive Transformer (NAT) is a family of text generation models, which aims to reduce the decoding latency by predicting the whole sentences in parallel.

Text Generation

Curriculum-Based Self-Training Makes Better Few-Shot Learners for Data-to-Text Generation

1 code implementation6 Jun 2022 Pei Ke, Haozhe Ji, Zhenyu Yang, Yi Huang, Junlan Feng, Xiaoyan Zhu, Minlie Huang

Despite the success of text-to-text pre-trained models in various natural language generation (NLG) tasks, the generation performance is largely restricted by the number of labeled data in downstream tasks, particularly in data-to-text generation tasks.

Data-to-Text Generation Unsupervised Pre-training

Many-Class Text Classification with Matching

no code implementations23 May 2022 Yi Song, Yuxian Gu, Minlie Huang

In this work, we formulate \textbf{T}ext \textbf{C}lassification as a \textbf{M}atching problem between the text and the labels, and propose a simple yet effective framework named TCM.

text-classification Text Classification

Directed Acyclic Transformer for Non-Autoregressive Machine Translation

1 code implementation16 May 2022 Fei Huang, Hao Zhou, Yang Liu, Hang Li, Minlie Huang

Non-autoregressive Transformers (NATs) significantly reduce the decoding latency by generating all tokens in parallel.

Knowledge Distillation Machine Translation +1

LaMemo: Language Modeling with Look-Ahead Memory

1 code implementation NAACL 2022 Haozhe Ji, Rongsheng Zhang, Zhenyu Yang, Zhipeng Hu, Minlie Huang

Although Transformers with fully connected self-attentions are powerful to model long-term dependencies, they are struggling to scale to long texts with thousands of words in language modeling.

Language Modelling

Chat-Capsule: A Hierarchical Capsule for Dialog-level Emotion Analysis

no code implementations23 Mar 2022 Yequan Wang, Xuying Meng, Yiyi Liu, Aixin Sun, Yao Wang, Yinhe Zheng, Minlie Huang

These models hence are not optimized for dialog-level emotion detection, i. e. to predict the emotion category of a dialog as a whole.

Emotion Recognition

Continual Prompt Tuning for Dialog State Tracking

1 code implementation ACL 2022 Qi Zhu, Bing Li, Fei Mi, Xiaoyan Zhu, Minlie Huang

A desirable dialog system should be able to continually learn new skills without forgetting old ones, and thereby adapt to new domains or tasks in its life cycle.

Continual Learning dialog state tracking +1

Acceleration of Federated Learning with Alleviated Forgetting in Local Training

1 code implementation ICLR 2022 Chencheng Xu, Zhiwei Hong, Minlie Huang, Tao Jiang

Here, we propose FedReg, an algorithm to accelerate FL with alleviated knowledge forgetting in the local training stage by regularizing locally trained parameters with the loss on generated pseudo data, which encode the knowledge of previous training data learned by the global model.

Distributed Optimization Federated Learning

Rethinking and Refining the Distinct Metric

1 code implementation ACL 2022 Siyang Liu, Sahand Sabour, Yinhe Zheng, Pei Ke, Xiaoyan Zhu, Minlie Huang

We provide both empirical and theoretical evidence to show that our method effectively removes the biases existing in the original distinct score.

Text Generation

AugESC: Dialogue Augmentation with Large Language Models for Emotional Support Conversation

1 code implementation26 Feb 2022 Chujie Zheng, Sahand Sabour, Jiaxin Wen, Zheng Zhang, Minlie Huang

Applying this approach, we construct AugESC, an augmented dataset for the ESC task, which largely extends the scale and topic coverage of the crowdsourced ESConv corpus.

Data Augmentation Dialogue Generation +2

Towards Identifying Social Bias in Dialog Systems: Frame, Datasets, and Benchmarks

1 code implementation16 Feb 2022 Jingyan Zhou, Jiawen Deng, Fei Mi, Yitong Li, Yasheng Wang, Minlie Huang, Xin Jiang, Qun Liu, Helen Meng

The research of open-domain dialog systems has been greatly prospered by neural models trained on large-scale corpora, however, such corpora often introduce various safety problems (e. g., offensive languages, biases, and toxic behaviors) that significantly hinder the deployment of dialog systems in practice.

Bias Detection Open-Domain Dialog

Youling: an AI-Assisted Lyrics Creation System

no code implementations EMNLP 2020 Rongsheng Zhang, Xiaoxi Mao, Le Li, Lin Jiang, Lin Chen, Zhiwei Hu, Yadong Xi, Changjie Fan, Minlie Huang

In the lyrics generation process, \textit{Youling} supports traditional one pass full-text generation mode as well as an interactive generation mode, which allows users to select the satisfactory sentences from generated candidates conditioned on preceding context.

Text Generation

COLD: A Benchmark for Chinese Offensive Language Detection

1 code implementation16 Jan 2022 Jiawen Deng, Jingyan Zhou, Hao Sun, Chujie Zheng, Fei Mi, Helen Meng, Minlie Huang

To this end, we propose a benchmark --COLD for Chinese offensive language analysis, including a Chinese Offensive Language Dataset --COLDATASET and a baseline detector --COLDETECTOR which is trained on the dataset.

Unsupervised Domain Adaptation with Adapter

no code implementations1 Nov 2021 Rongsheng Zhang, Yinhe Zheng, Xiaoxi Mao, Minlie Huang

However, fine-tuning all the parameters of the PrLM on a small domain-specific corpus distort the learned generic knowledge, and it is also expensive to deployment a whole fine-tuned PrLM for each domain.

Unsupervised Domain Adaptation

On the Safety of Conversational Models: Taxonomy, Dataset, and Benchmark

1 code implementation Findings (ACL) 2022 Hao Sun, Guangxuan Xu, Jiawen Deng, Jiale Cheng, Chujie Zheng, Hao Zhou, Nanyun Peng, Xiaoyan Zhu, Minlie Huang

We propose a taxonomy for dialogue safety specifically designed to capture unsafe behaviors in human-bot dialogue settings, with focuses on context-sensitive unsafety, which is under-explored in prior works.

Answering Open-Domain Multi-Answer Questions via a Recall-then-Verify Framework

1 code implementation ACL 2022 Zhihong Shao, Minlie Huang

Open-domain questions are likely to be open-ended and ambiguous, leading to multiple valid answers.

valid

DiscoDVT: Generating Long Text with Discourse-Aware Discrete Variational Transformer

1 code implementation EMNLP 2021 Haozhe Ji, Minlie Huang

Despite the recent advances in applying pre-trained language models to generate high-quality texts, generating long passages that maintain long-range coherence is yet challenging for these models.

Story Generation

Exploring Prompt-based Few-shot Learning for Grounded Dialog Generation

no code implementations14 Sep 2021 Chujie Zheng, Minlie Huang

In this work, we focus on the few-shot learning for grounded dialog generation (GDG).

Few-Shot Learning

CEM: Commonsense-aware Empathetic Response Generation

1 code implementation13 Sep 2021 Sahand Sabour, Chujie Zheng, Minlie Huang

We evaluate our approach on EmpatheticDialogues, which is a widely-used benchmark dataset for empathetic response generation.

Empathetic Response Generation Response Generation

PPT: Pre-trained Prompt Tuning for Few-shot Learning

1 code implementation ACL 2022 Yuxian Gu, Xu Han, Zhiyuan Liu, Minlie Huang

To ensure the generalization of PPT, we formulate similar classification tasks into a unified task form and pre-train soft prompts for this unified task.

Attribute Few-Shot Learning

LOT: A Story-Centric Benchmark for Evaluating Chinese Long Text Understanding and Generation

2 code implementations30 Aug 2021 Jian Guan, Zhuoer Feng, Yamei Chen, Ruilin He, Xiaoxi Mao, Changjie Fan, Minlie Huang

Therefore, we propose a story-centric benchmark named LOT for evaluating Chinese long text modeling, which aggregates two understanding tasks and two generation tasks.

Decoder Text Infilling

Self-training Improves Pre-training for Few-shot Learning in Task-oriented Dialog Systems

1 code implementation EMNLP 2021 Fei Mi, Wanhao Zhou, Fengyu Cai, Lingjing Kong, Minlie Huang, Boi Faltings

In this paper, we devise a self-training approach to utilize the abundant unlabeled dialog data to further improve state-of-the-art pre-trained models in few-shot learning scenarios for ToD systems.

dialog state tracking Few-Shot Learning +4

EVA: An Open-Domain Chinese Dialogue System with Large-Scale Generative Pre-Training

2 code implementations3 Aug 2021 Hao Zhou, Pei Ke, Zheng Zhang, Yuxian Gu, Yinhe Zheng, Chujie Zheng, Yida Wang, Chen Henry Wu, Hao Sun, Xiaocong Yang, Bosi Wen, Xiaoyan Zhu, Minlie Huang, Jie Tang

Although pre-trained language models have remarkably enhanced the generation ability of dialogue systems, open-domain Chinese dialogue systems are still limited by the dialogue data and the model size compared with English ones.

KuiLeiXi: a Chinese Open-Ended Text Adventure Game

no code implementations ACL 2021 Yadong Xi, Xiaoxi Mao, Le Li, Lei Lin, Yanjiang Chen, Shuhan Yang, Xuhan Chen, Kailun Tao, Zhi Li, Gongzheng li, Lin Jiang, Siyan Liu, Zeng Zhao, Minlie Huang, Changjie Fan, Zhipeng Hu

Equipped with GPT-2 and the latest GPT-3, AI Dungeon has been seen as a famous example of the powerful text generation capabilities of large-scale pre-trained language models, and a possibility for future games.

Story Generation

MConv: An Environment for Multimodal Conversational Search across Multiple Domains

1 code implementation SIGIR 2021 Lizi Liao, Le Hong Long, Zheng Zhang, Minlie Huang, Tat-Seng Chua

Second, a set of benchmark results for dialogue state tracking, conversational recommendation, response generation as well as a unified model for multiple tasks are reported.

Conversational Search Dialogue State Tracking +1

End-to-End Task-Oriented Dialog Modeling with Semi-Structured Knowledge Management

1 code implementation22 Jun 2021 Silin Gao, Ryuichi Takanobu, Antoine Bosselut, Minlie Huang

To address this task, we propose a TOD system with semi-structured knowledge management, SeKnow, which extends the belief state to manage knowledge with both structured and unstructured contents.

Language Modelling Management

CPM-2: Large-scale Cost-effective Pre-trained Language Models

2 code implementations20 Jun 2021 Zhengyan Zhang, Yuxian Gu, Xu Han, Shengqi Chen, Chaojun Xiao, Zhenbo Sun, Yuan YAO, Fanchao Qi, Jian Guan, Pei Ke, Yanzheng Cai, Guoyang Zeng, Zhixing Tan, Zhiyuan Liu, Minlie Huang, Wentao Han, Yang Liu, Xiaoyan Zhu, Maosong Sun

We present a suite of cost-effective techniques for the use of PLMs to deal with the efficiency issues of pre-training, fine-tuning, and inference.

Decoder

JointGT: Graph-Text Joint Representation Learning for Text Generation from Knowledge Graphs

1 code implementation Findings (ACL) 2021 Pei Ke, Haozhe Ji, Yu Ran, Xin Cui, LiWei Wang, Linfeng Song, Xiaoyan Zhu, Minlie Huang

Existing pre-trained models for knowledge-graph-to-text (KG-to-text) generation simply fine-tune text-to-text pre-trained models such as BART or T5 on KG-to-text datasets, which largely ignore the graph structure during encoding and lack elaborate pre-training tasks to explicitly model graph-text alignments.

Graph Reconstruction KG-to-Text Generation +3

A Mutual Information Maximization Approach for the Spurious Solution Problem in Weakly Supervised Question Answering

1 code implementation ACL 2021 Zhihong Shao, Lifeng Shang, Qun Liu, Minlie Huang

This setting gives rise to the spurious solution problem: there may exist many spurious solutions that coincidentally derive the correct answer, but training on such solutions can hurt model performance (e. g., producing wrong solutions or answers).

Question Answering

Semantic-Enhanced Explainable Finetuning for Open-Domain Dialogues

no code implementations6 Jun 2021 Yinhe Zheng, Yida Wang, Pei Ke, Zhenyu Yang, Minlie Huang

This paper propose to combine pretrained language models with the modular dialogue paradigm for open-domain dialogue modeling.

Informativeness Language Modelling +1

NAST: A Non-Autoregressive Generator with Word Alignment for Unsupervised Text Style Transfer

1 code implementation Findings (ACL) 2021 Fei Huang, Zikai Chen, Chen Henry Wu, Qihan Guo, Xiaoyan Zhu, Minlie Huang

First, we observe that most words in the transferred sentence can be aligned with related words in the source sentence, so we explicitly model word alignments to suppress irrelevant words.

Sentence Style Transfer +3

PsyQA: A Chinese Dataset for Generating Long Counseling Text for Mental Health Support

2 code implementations Findings (ACL) 2021 Hao Sun, Zhenru Lin, Chujie Zheng, Siyang Liu, Minlie Huang

In this paper, we propose PsyQA, a Chinese dataset of psychological health support in the form of question and answer pair.

Towards Emotional Support Dialog Systems

1 code implementation ACL 2021 Siyang Liu, Chujie Zheng, Orianna Demasi, Sahand Sabour, Yu Li, Zhou Yu, Yong Jiang, Minlie Huang

Emotional support is a crucial ability for many conversation scenarios, including social interactions, mental health support, and customer service chats.

Diversifying Dialog Generation via Adaptive Label Smoothing

1 code implementation ACL 2021 Yida Wang, Yinhe Zheng, Yong Jiang, Minlie Huang

Neural dialogue generation models trained with the one-hot target distribution suffer from the over-confidence issue, which leads to poor generation diversity as widely reported in the literature.

Decoder Dialogue Generation

OpenMEVA: A Benchmark for Evaluating Open-ended Story Generation Metrics

1 code implementation ACL 2021 Jian Guan, Zhexin Zhang, Zhuoer Feng, Zitao Liu, Wenbiao Ding, Xiaoxi Mao, Changjie Fan, Minlie Huang

Automatic metrics are essential for developing natural language generation (NLG) models, particularly for open-ended language generation tasks such as story generation.

Story Generation

Long Text Generation by Modeling Sentence-Level and Discourse-Level Coherence

1 code implementation ACL 2021 Jian Guan, Xiaoxi Mao, Changjie Fan, Zitao Liu, Wenbiao Ding, Minlie Huang

Generating long and coherent text is an important but challenging task, particularly for open-ended language generation tasks such as story generation.

Decoder Semantic Similarity +3

Stylized Story Generation with Style-Guided Planning

no code implementations Findings (ACL) 2021 Xiangzhe Kong, Jialiang Huang, Ziquan Tung, Jian Guan, Minlie Huang

Current storytelling systems focus more ongenerating stories with coherent plots regard-less of the narration style, which is impor-tant for controllable text generation.

Story Generation

CoMAE: A Multi-factor Hierarchical Framework for Empathetic Response Generation

1 code implementation Findings (ACL) 2021 Chujie Zheng, Yong liu, Wei Chen, Yongcai Leng, Minlie Huang

However, existing methods for empathetic response generation usually either consider only one empathy factor or ignore the hierarchical relationships between different factors, leading to a weak ability of empathy modeling.

Empathetic Response Generation Open-Domain Dialog +1

A Text GAN for Language Generation with Non-Autoregressive Generator

no code implementations1 Jan 2021 Fei Huang, Jian Guan, Pei Ke, Qihan Guo, Xiaoyan Zhu, Minlie Huang

Despite the great success of Generative Adversarial Networks (GANs) in generating high-quality images, GANs for text generation still face two major challenges: first, most text GANs are unstable in training mainly due to ineffective optimization of the generator, and they heavily rely on maximum likelihood pretraining; second, most text GANs adopt autoregressive generators without latent variables, which largely limits the ability to learn latent representations for natural language text.

Decipherment Representation Learning +2

Robustness Testing of Language Understanding in Task-Oriented Dialog

2 code implementations ACL 2021 Jiexi Liu, Ryuichi Takanobu, Jiaxin Wen, Dazhen Wan, Hongguang Li, Weiran Nie, Cheng Li, Wei Peng, Minlie Huang

Most language understanding models in task-oriented dialog systems are trained on a small amount of annotated training data, and evaluated in a small set from the same distribution.

Data Augmentation Natural Language Understanding

AdvExpander: Generating Natural Language Adversarial Examples by Expanding Text

no code implementations18 Dec 2020 Zhihong Shao, Zitao Liu, Jiyong Zhang, Zhongqin Wu, Minlie Huang

In this paper, we present AdvExpander, a method that crafts new adversarial examples by expanding text, which is complementary to previous substitution-based methods.

Text Matching

Reinforced Molecular Optimization with Neighborhood-Controlled Grammars

1 code implementation NeurIPS 2020 Chencheng Xu, Qiao Liu, Minlie Huang, Tao Jiang

A major challenge in the pharmaceutical industry is to design novel molecules with specific desired properties, especially when the property evaluation is costly.

 Ranked #1 on Molecular Graph Generation on ZINC (QED Top-3 metric)

Graph Generation Molecular Graph Generation

CR-Walker: Tree-Structured Graph Reasoning and Dialog Acts for Conversational Recommendation

1 code implementation EMNLP 2021 Wenchang Ma, Ryuichi Takanobu, Minlie Huang

Growing interests have been attracted in Conversational Recommender Systems (CRS), which explore user preference through conversational interactions in order to make appropriate recommendation.

Recommendation Systems Response Generation +1

MultiWOZ 2.3: A multi-domain task-oriented dialogue dataset enhanced with annotation corrections and co-reference annotation

3 code implementations12 Oct 2020 Ting Han, Ximing Liu, Ryuichi Takanobu, Yixin Lian, Chongxuan Huang, Dazhen Wan, Wei Peng, Minlie Huang

In this paper, we introduce MultiWOZ 2. 3, in which we differentiate incorrect annotations in dialogue acts from dialogue states, identifying a lack of co-reference when publishing the updated dataset.

Dialogue State Tracking Natural Language Understanding +1

Stylized Dialogue Response Generation Using Stylized Unpaired Texts

1 code implementation27 Sep 2020 Yinhe Zheng, Zikai Chen, Rongsheng Zhang, Shilei Huang, Xiaoxi Mao, Minlie Huang

However, this task is far from well-explored due to the difficulties of rendering a particular style in coherent responses, especially when the target style is embedded only in unpaired texts that cannot be directly used to train the dialogue model.

Decoder Dialogue Generation +1

Language Generation with Multi-Hop Reasoning on Commonsense Knowledge Graph

1 code implementation EMNLP 2020 Haozhe Ji, Pei Ke, Shaohan Huang, Furu Wei, Xiaoyan Zhu, Minlie Huang

Despite the success of generative pre-trained language models on a series of text generation tasks, they still suffer in cases where reasoning over underlying commonsense knowledge is required during generation.

Text Generation

Generating Commonsense Explanation by Extracting Bridge Concepts from Reasoning Paths

no code implementations Asian Chapter of the Association for Computational Linguistics 2020 Haozhe Ji, Pei Ke, Shaohan Huang, Furu Wei, Minlie Huang

Commonsense explanation generation aims to empower the machine's sense-making capability by generating plausible explanations to statements against commonsense.

Explanation Generation

Dialogue Distillation: Open-Domain Dialogue Augmentation Using Unpaired Data

1 code implementation EMNLP 2020 Rongsheng Zhang, Yinhe Zheng, Jianzhi Shao, Xiaoxi Mao, Yadong Xi, Minlie Huang

Further, a model-level distillation process is employed to distill a teacher model trained on high-quality paired data to augmented dialogue pairs, thereby preventing dialogue models from being affected by the noise in the augmented data.

Data Augmentation

Difference-aware Knowledge Selection for Knowledge-grounded Conversation Generation

1 code implementation Findings of the Association for Computational Linguistics 2020 Chujie Zheng, Yunbo Cao, Daxin Jiang, Minlie Huang

In a multi-turn knowledge-grounded dialog, the difference between the knowledge selected at different turns usually provides potential clues to knowledge selection, which has been largely neglected in previous research.

UNION: An Unreferenced Metric for Evaluating Open-ended Story Generation

1 code implementation EMNLP 2020 Jian Guan, Minlie Huang

Experiments on two story datasets demonstrate that UNION is a reliable measure for evaluating the quality of generated stories, which correlates better with human judgments and is more generalizable than existing state-of-the-art metrics.

Story Generation

A Large-Scale Chinese Short-Text Conversation Dataset

2 code implementations10 Aug 2020 Yida Wang, Pei Ke, Yinhe Zheng, Kaili Huang, Yong Jiang, Xiaoyan Zhu, Minlie Huang

The cleaned dataset and the pre-training models will facilitate the research of short-text conversation modeling.

Dialogue Generation Short-Text Conversation

FIVES: Feature Interaction Via Edge Search for Large-Scale Tabular Data

no code implementations29 Jul 2020 Yuexiang Xie, Zhen Wang, Yaliang Li, Bolin Ding, Nezihe Merve Gürel, Ce Zhang, Minlie Huang, Wei. Lin, Jingren Zhou

Then we instantiate this search strategy by optimizing both a dedicated graph neural network (GNN) and the adjacency tensor associated with the defined feature graph.

Graph Neural Network Recommendation Systems

Knowledge-Aided Open-Domain Question Answering

no code implementations9 Jun 2020 Mantong Zhou, Zhouxing Shi, Minlie Huang, Xiaoyan Zhu

During document retrieval, a candidate document is scored by considering its relationship to the question and other documents.

Open-Domain Question Answering Reading Comprehension +1

A Self-Training Method for Machine Reading Comprehension with Soft Evidence Extraction

1 code implementation ACL 2020 Yilin Niu, Fangkai Jiao, Mantong Zhou, Ting Yao, Jingfang Xu, Minlie Huang

Neural models have achieved great success on machine reading comprehension (MRC), many of which typically consist of two components: an evidence extractor and an answer predictor.

Machine Reading Comprehension Multi-Choice MRC +1

Learning Goal-oriented Dialogue Policy with Opposite Agent Awareness

no code implementations Asian Chapter of the Association for Computational Linguistics 2020 Zheng Zhang, Lizi Liao, Xiaoyan Zhu, Tat-Seng Chua, Zitao Liu, Yan Huang, Minlie Huang

Most existing approaches for goal-oriented dialogue policy learning used reinforcement learning, which focuses on the target agent policy and simply treat the opposite agent policy as part of the environment.

Decision Making

Multi-Agent Task-Oriented Dialog Policy Learning with Role-Aware Reward Decomposition

1 code implementation ACL 2020 Ryuichi Takanobu, Runze Liang, Minlie Huang

To avoid explicitly building a user simulator beforehand, we propose Multi-Agent Dialog Policy Learning, which regards both the system and the user as the dialog agents.

reinforcement-learning Reinforcement Learning (RL)

KdConv: A Chinese Multi-domain Dialogue Dataset Towards Multi-turn Knowledge-driven Conversation

1 code implementation ACL 2020 Hao Zhou, Chujie Zheng, Kaili Huang, Minlie Huang, Xiaoyan Zhu

The research of knowledge-driven conversational systems is largely limited due to the lack of dialog data which consist of multi-turn conversations on multiple topics and with knowledge annotations.

Domain Adaptation Knowledge Graphs +1

Recent Advances and Challenges in Task-oriented Dialog System

no code implementations17 Mar 2020 Zheng Zhang, Ryuichi Takanobu, Qi Zhu, Minlie Huang, Xiaoyan Zhu

Due to the significance and value in human-computer interaction and natural language processing, task-oriented dialog systems are attracting more and more attention in both academic and industrial communities.

Automatic Perturbation Analysis for Scalable Certified Robustness and Beyond

5 code implementations NeurIPS 2020 Kaidi Xu, Zhouxing Shi, huan zhang, Yihan Wang, Kai-Wei Chang, Minlie Huang, Bhavya Kailkhura, Xue Lin, Cho-Jui Hsieh

Linear relaxation based perturbation analysis (LiRPA) for neural networks, which computes provable linear bounds of output neurons given a certain amount of input perturbation, has become a core component in robustness verification and certified defense.

Quantization

CrossWOZ: A Large-Scale Chinese Cross-Domain Task-Oriented Dialogue Dataset

2 code implementations TACL 2020 Qi Zhu, Kaili Huang, Zheng Zhang, Xiaoyan Zhu, Minlie Huang

To advance multi-domain (cross-domain) dialogue modeling as well as alleviate the shortage of Chinese task-oriented datasets, we propose CrossWOZ, the first large-scale Chinese Cross-Domain Wizard-of-Oz task-oriented dataset.

Dialogue State Tracking Task-Oriented Dialogue Systems +1

Robustness Verification for Transformers

1 code implementation ICLR 2020 Zhouxing Shi, huan zhang, Kai-Wei Chang, Minlie Huang, Cho-Jui Hsieh

Robustness verification that aims to formally certify the prediction behavior of neural networks has become an important tool for understanding model behavior and obtaining safety guarantees.

Position Sentiment Analysis

ConvLab-2: An Open-Source Toolkit for Building, Evaluating, and Diagnosing Dialogue Systems

1 code implementation ACL 2020 Qi Zhu, Zheng Zhang, Yan Fang, Xiang Li, Ryuichi Takanobu, Jinchao Li, Baolin Peng, Jianfeng Gao, Xiaoyan Zhu, Minlie Huang

We present ConvLab-2, an open-source toolkit that enables researchers to build task-oriented dialogue systems with state-of-the-art models, perform an end-to-end evaluation, and diagnose the weakness of systems.

Task-Oriented Dialogue Systems

CoTK: An Open-Source Toolkit for Fast Development and Fair Evaluation of Text Generation

1 code implementation3 Feb 2020 Fei Huang, Dazhen Wan, Zhihong Shao, Pei Ke, Jian Guan, Yilin Niu, Xiaoyan Zhu, Minlie Huang

In text generation evaluation, many practical issues, such as inconsistent experimental settings and metric implementations, are often ignored but lead to unfair evaluation and untenable conclusions.

Text Generation

A Knowledge-Enhanced Pretraining Model for Commonsense Story Generation

1 code implementation TACL 2020 Jian Guan, Fei Huang, Zhihao Zhao, Xiaoyan Zhu, Minlie Huang

To further capture the causal and temporal dependencies between the sentences in a reasonable story, we employ multi-task learning which combines a discriminative objective to distinguish true and fake stories during fine-tuning.

Multi-Task Learning Story Generation

Robust Reading Comprehension with Linguistic Constraints via Posterior Regularization

no code implementations16 Nov 2019 Mantong Zhou, Minlie Huang, Xiaoyan Zhu

In this paper, we address the over-confidence issue and the over-sensitivity issue existing in current RC models simultaneously with the help of external linguistic knowledge.

Machine Reading Comprehension

A Pre-training Based Personalized Dialogue Generation Model with Persona-sparse Data

2 code implementations12 Nov 2019 Yinhe Zheng, Rongsheng Zhang, Xiaoxi Mao, Minlie Huang

Further, to incorporate the target persona in the decoding process and to balance its contribution, an attention routing structure is devised in the decoder to merge features extracted from the target persona and dialogue contexts using dynamically predicted weights.

Attribute Decoder +2

SentiLARE: Sentiment-Aware Language Representation Learning with Linguistic Knowledge

1 code implementation EMNLP 2020 Pei Ke, Haozhe Ji, Siyang Liu, Xiaoyan Zhu, Minlie Huang

To benefit the downstream tasks in sentiment analysis, we propose a novel language representation model called SentiLARE, which introduces word-level linguistic knowledge including part-of-speech tag and sentiment po