1 code implementation • 26 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.
2 code implementations • 1 Feb 2024 • Haozhe Ji, Cheng Lu, Yilin Niu, Pei Ke, Hongning Wang, Jun Zhu, Jie Tang, Minlie Huang
We prove that EXO is guaranteed to optimize in the same direction as the RL algorithms asymptotically for arbitary parametrization of the policy, while enables efficient optimization by circumventing the complexities associated with RL algorithms.
2 code implementations • 30 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.
1 code implementation • 30 Nov 2023 • Xiao Liu, Xuanyu Lei, Shengyuan Wang, Yue Huang, Zhuoer Feng, Bosi Wen, Jiale Cheng, Pei Ke, Yifan Xu, Weng Lam Tam, Xiaohan Zhang, Lichao Sun, Hongning Wang, Jing Zhang, Minlie Huang, Yuxiao Dong, Jie Tang
We will provide public APIs for evaluating AlignBench with CritiqueLLM to facilitate the evaluation of LLMs' Chinese alignment.
1 code implementation • 29 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.
1 code implementation • 15 Nov 2023 • Zhexin Zhang, Junxiao Yang, Pei Ke, Minlie Huang
We hope our work could contribute to the comprehension of jailbreaking attacks and defenses, and shed light on the relationship between LLMs' capability and safety.
1 code implementation • 7 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.
no code implementations • 2 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.
1 code implementation • 13 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.
1 code implementation • 6 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.
1 code implementation • 24 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.
1 code implementation • 26 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.
1 code implementation • 17 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.
1 code implementation • 6 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.
1 code implementation • ACL 2022 • Pei Ke, Hao Zhou, Yankai Lin, Peng Li, Jie zhou, Xiaoyan Zhu, Minlie Huang
Existing reference-free metrics have obvious limitations for evaluating controlled text generation models.
1 code implementation • 17 Mar 2022 • Yuxian Gu, Jiaxin Wen, Hao Sun, Yi Song, Pei Ke, Chujie Zheng, Zheng Zhang, Jianzhu Yao, Lei Liu, Xiaoyan Zhu, Minlie Huang
Large-scale pre-training has shown remarkable performance in building open-domain dialogue systems.
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.
2 code implementations • 3 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.
2 code implementations • 20 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.
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.
Ranked #1 on KG-to-Text Generation on WebQuestions
no code implementations • 6 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.
no code implementations • 1 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.
6 code implementations • 1 Dec 2020 • Zhengyan Zhang, Xu Han, Hao Zhou, Pei Ke, Yuxian Gu, Deming Ye, Yujia Qin, Yusheng Su, Haozhe Ji, Jian Guan, Fanchao Qi, Xiaozhi Wang, Yanan Zheng, Guoyang Zeng, Huanqi Cao, Shengqi Chen, Daixuan Li, Zhenbo Sun, Zhiyuan Liu, Minlie Huang, Wentao Han, Jie Tang, Juanzi Li, Xiaoyan Zhu, Maosong Sun
However, applying GPT-3 to address Chinese NLP tasks is still challenging, as the training corpus of GPT-3 is primarily English, and the parameters are not publicly available.
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.
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.
2 code implementations • 10 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.
1 code implementation • 3 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.
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 polarity (inferred from SentiWordNet) into pre-trained models.
1 code implementation • IJCNLP 2019 • Pei Ke, Fei Huang, Minlie Huang, Xiaoyan Zhu
The generator is optimized with maximum likelihood estimation augmented by the discriminator's rewards instead of policy gradient.
1 code implementation • ACL 2018 • Pei Ke, Jian Guan, Minlie Huang, Xiaoyan Zhu
Experiments show that our model outperforms state-of-the-art baselines, and it has the ability to generate responses with both controlled sentence function and informative content.