1 code implementation • 6 Jun 2025 • Sooyung Choi, JaeHyeok Lee, Xiaoyuan Yi, Jing Yao, Xing Xie, JinYeong Bak
Using a dataset with detailed safety categories, we find significant correlations between value alignment and safety risks, supported by psychological hypotheses.
no code implementations • 18 May 2025 • Shitong Duan, Xiaoyuan Yi, Peng Zhang, Dongkuan Xu, Jing Yao, Tun Lu, Ning Gu, Xing Xie
Assessing Large Language Models (LLMs)' underlying value differences enables comprehensive comparison of their misalignment, cultural adaptability, and biases.
1 code implementation • 26 Mar 2025 • Huanhuan Ma, Haisong Gong, Xiaoyuan Yi, Xing Xie, Dongkuan Xu
Through extensive experiments, we demonstrate that: 1) CSI effectively captures nuanced emotional patterns, revealing significant variation in LLMs across languages and contexts; 2) Compared to current approaches, CSI significantly improves reliability, yielding more consistent results; and 3) The correlation between CSI scores and the sentiment of LLM's real-world outputs exceeds 0. 85, demonstrating its strong validity in predicting LLM behavior.
no code implementations • 8 Mar 2025 • HyunJin Kim, Xiaoyuan Yi, Jing Yao, Muhua Huang, JinYeong Bak, James Evans, Xing Xie
The recent leap in AI capabilities, driven by big generative models, has sparked the possibility of achieving Artificial General Intelligence (AGI) and further triggered discussions on Artificial Superintelligence (ASI), a system surpassing all humans across all domains.
2 code implementations • 24 Feb 2025 • Zhenghao Liu, Xingsheng Zhu, Tianshuo Zhou, Xinyi Zhang, Xiaoyuan Yi, Yukun Yan, Yu Gu, Ge Yu, Maosong Sun
This paper introduces Multi-Modal Retrieval-Augmented Generation (M^2RAG), a benchmark designed to evaluate the effectiveness of Multi-modal Large Language Models (MLLMs) in leveraging knowledge from multi-modal retrieval documents.
1 code implementation • 21 Feb 2025 • Pengcheng Huang, Zhenghao Liu, Yukun Yan, Xiaoyuan Yi, Hao Chen, Zhiyuan Liu, Maosong Sun, Tong Xiao, Ge Yu, Chenyan Xiong
Knowledge-Augmented Generation (KAG) has shown great promise in updating the internal memory of Large Language Models (LLMs) by integrating external knowledge.
no code implementations • 13 Jan 2025 • Jing Yao, Xiaoyuan Yi, Shitong Duan, Jindong Wang, Yuzhuo Bai, Muhua Huang, Peng Zhang, Tun Lu, Zhicheng Dou, Maosong Sun, Xing Xie
As Large Language Models (LLMs) achieve remarkable breakthroughs, aligning their values with humans has become imperative for their responsible development and customized applications.
no code implementations • 21 Dec 2024 • HyunJin Kim, Xiaoyuan Yi, Jing Yao, Jianxun Lian, Muhua Huang, Shitong Duan, JinYeong Bak, Xing Xie
The emergence of large language models (LLMs) has sparked the possibility of about Artificial Superintelligence (ASI), a hypothetical AI system surpassing human intelligence.
1 code implementation • 16 Oct 2024 • Xingqi Wang, Xiaoyuan Yi, Xing Xie, Jia Jia
To optimize the value encoder, we also develop a framework to automatically construct a text-image preference dataset of 86k (prompt, aligned image, violating image, value principle) samples.
no code implementations • 26 Sep 2024 • Yan Liu, Xiaoyuan Yi, Xiaokang Chen, Jing Yao, Jingwei Yi, Daoguang Zan, Zheng Liu, Xing Xie, Tsung-Yi Ho
Despite the vital role reward models play in alignment, previous works have consistently overlooked their performance and used off-the-shelf reward models arbitrarily without verification, rendering the reward model ``\emph{an elephant in the room}''.
no code implementations • 15 Jul 2024 • Jing Yao, Xiaoyuan Yi, Xing Xie
The rapid progress in Large Language Models (LLMs) poses potential risks such as generating unethical content.
no code implementations • 20 Jun 2024 • Han Jiang, Xiaoyuan Yi, Zhihua Wei, Ziang Xiao, Shu Wang, Xing Xie
Unlike traditional adaptive testing methods that rely on a static test item pool, GETA probes the underlying moral boundaries of LLMs by dynamically generating test items tailored to model capability.
1 code implementation • 17 May 2024 • Yuqing Lan, Zhenghao Liu, Yu Gu, Xiaoyuan Yi, Xiaohua LI, Liner Yang, Ge Yu
Nevertheless, the noisy nodes usually propagate their semantics via the edges of the reasoning graph, which misleads the semantic representations of other nodes and amplifies the noise signals.
no code implementations • 19 Apr 2024 • Pablo Biedma, Xiaoyuan Yi, Linus Huang, Maosong Sun, Xing Xie
Recent advancements in Large Language Models (LLMs) have revolutionized the AI field but also pose potential safety and ethical risks.
no code implementations • 7 Mar 2024 • Xinpeng Wang, Shitong Duan, Xiaoyuan Yi, Jing Yao, Shanlin Zhou, Zhihua Wei, Peng Zhang, Dongkuan Xu, Maosong Sun, Xing Xie
Big models have achieved revolutionary breakthroughs in the field of AI, but they might also pose potential concerns.
1 code implementation • 6 Mar 2024 • Shitong Duan, Xiaoyuan Yi, Peng Zhang, Yan Liu, Zheng Liu, Tun Lu, Xing Xie, Ning Gu
Large language models (LLMs) have revolutionized the role of AI, yet pose potential social risks.
no code implementations • 29 Feb 2024 • Xukun Liu, Zhiyuan Peng, Xiaoyuan Yi, Xing Xie, Lirong Xiang, Yuchen Liu, Dongkuan Xu
While achieving remarkable progress in a broad range of tasks, large language models (LLMs) remain significantly limited in properly using massive external tools.
1 code implementation • 27 Jan 2024 • Buqiang Xu, Xin Dai, Zhenghao Liu, Huiyuan Xie, Xiaoyuan Yi, Shuo Wang, Yukun Yan, Liner Yang, Yu Gu, Ge Yu
In this paper, we propose LegalDuet, which continuously pretrains language models to learn a more tailored embedding space for representing legal cases.
1 code implementation • 13 Dec 2023 • Xinpeng Wang, Xiaoyuan Yi, Han Jiang, Shanlin Zhou, Zhihua Wei, Xing Xie
Warning: this paper includes model outputs showing offensive content.
1 code implementation • 28 Nov 2023 • Yuhang Wang, Yanxu Zhu, Chao Kong, Shuyu Wei, Xiaoyuan Yi, Xing Xie, Jitao Sang
This benchmark serves as a valuable resource for cultural studies in LLMs, paving the way for more culturally aware and sensitive models.
1 code implementation • 16 Nov 2023 • Jing Yao, Wei Xu, Jianxun Lian, Xiting Wang, Xiaoyuan Yi, Xing Xie
In this paper, we propose a general paradigm that augments LLMs with DOmain-specific KnowledgE to enhance their performance on practical applications, namely DOKE.
no code implementations • 15 Nov 2023 • Jing Yao, Xiaoyuan Yi, Xiting Wang, Yifan Gong, Xing Xie
The rapid advancement of Large Language Models (LLMs) has attracted much attention to value alignment for their responsible development.
no code implementations • 26 Oct 2023 • Xiaoyuan Yi, Jing Yao, Xiting Wang, Xing Xie
Big models have greatly advanced AI's ability to understand, generate, and manipulate information and content, enabling numerous applications.
no code implementations • 17 Oct 2023 • Shitong Duan, Xiaoyuan Yi, Peng Zhang, Tun Lu, Xing Xie, Ning Gu
We discovered that most models are essentially misaligned, necessitating further ethical value alignment.
no code implementations • 23 Aug 2023 • Jing Yao, Xiaoyuan Yi, Xiting Wang, Jindong Wang, Xing Xie
Big models, exemplified by Large Language Models (LLMs), are models typically pre-trained on massive data and comprised of enormous parameters, which not only obtain significantly improved performance across diverse tasks but also present emergent capabilities absent in smaller models.
1 code implementation • 6 Jul 2023 • Yupeng Chang, Xu Wang, Jindong Wang, Yuan Wu, Linyi Yang, Kaijie Zhu, Hao Chen, Xiaoyuan Yi, Cunxiang Wang, Yidong Wang, Wei Ye, Yue Zhang, Yi Chang, Philip S. Yu, Qiang Yang, Xing Xie
Large language models (LLMs) are gaining increasing popularity in both academia and industry, owing to their unprecedented performance in various applications.
1 code implementation • 17 Jun 2023 • Yuxi Feng, Xiaoyuan Yi, Laks V. S. Lakshmanan, Xing Xie
Self-training (ST) has come to fruition in language understanding tasks by producing pseudo labels, which reduces the labeling bottleneck of language model fine-tuning.
no code implementations • 19 May 2023 • Jinyi Hu, Xu Han, Xiaoyuan Yi, Yutong Chen, Wenhao Li, Zhiyuan Liu, Maosong Sun
IAP optimizes only a separate Chinese text encoder with all other parameters fixed to align Chinese semantics space to the English one in CLIP.
1 code implementation • 16 Dec 2022 • Yuxi Feng, Xiaoyuan Yi, Xiting Wang, Laks V. S. Lakshmanan, Xing Xie
Augmented by only self-generated pseudo text, generation models over-emphasize exploitation of the previously learned space, suffering from a constrained generalization boundary.
1 code implementation • 14 Nov 2022 • Wenhao Li, Xiaoyuan Yi, Jinyi Hu, Maosong Sun, Xing Xie
In this work, we dig into the intrinsic mechanism of this problem and found that sparser attention values in Transformer could improve diversity.
no code implementations • 22 Oct 2022 • Jinyi Hu, Xiaoyuan Yi, Wenhao Li, Maosong Sun, Xing Xie
We demonstrate that TRACE could enhance the entanglement of each segment and preceding latent variables and deduce a non-zero lower bound of the KL term, providing a theoretical guarantee of generation diversity.
1 code implementation • 13 Oct 2022 • Seungeon Lee, Xiting Wang, Sungwon Han, Xiaoyuan Yi, Xing Xie, Meeyoung Cha
We present SELOR, a framework for integrating self-explaining capabilities into a given deep model to achieve both high prediction performance and human precision.
no code implementations • 10 Oct 2022 • Zonghan Yang, Xiaoyuan Yi, Peng Li, Yang Liu, Xing Xie
Warning: this paper contains model outputs exhibiting offensiveness and biases.
1 code implementation • NAACL 2022 • Jinyi Hu, Xiaoyuan Yi, Wenhao Li, Maosong Sun, Xing Xie
The past several years have witnessed Variational Auto-Encoder's superiority in various text generation tasks.
1 code implementation • 3 Jun 2021 • Wenhao Li, Fanchao Qi, Maosong Sun, Xiaoyuan Yi, Jiarui Zhang
We hope this dataset can further enhance the study on incorporating deep semantics into the understanding and generation system of Chinese classical poetry.
1 code implementation • NAACL 2021 • Zhenghao Liu, Xiaoyuan Yi, Maosong Sun, Liner Yang, Tat-Seng Chua
Grammatical Error Correction (GEC) aims to correct writing errors and help language learners improve their writing skills.
Ranked #1 on
Grammatical Error Detection
on FCE
no code implementations • 13 Mar 2020 • Xiaoyuan Yi, Ruoyu Li, Cheng Yang, Wenhao Li, Maosong Sun
Though recent neural models make prominent progress in some criteria of poetry quality, generated poems still suffer from the problem of poor diversity.
no code implementations • ACL 2019 • Guo Zhipeng, Xiaoyuan Yi, Maosong Sun, Wenhao Li, Cheng Yang, Jiannan Liang, Huimin Chen, Yuhui Zhang, Ruoyu Li
By exposing the options of poetry genres, styles and revision modes, Jiuge, acting as a professional assistant, allows constant and active participation of users in poetic creation.
no code implementations • EMNLP 2018 • Cheng Yang, Maosong Sun, Xiaoyuan Yi, Wenhao Li
The ability to write diverse poems in different styles under the same poetic imagery is an important characteristic of human poetry writing.
no code implementations • EMNLP 2018 • Xiaoyuan Yi, Maosong Sun, Ruoyu Li, Wenhao Li
Human experts evaluate poetry in terms of some specific criteria, instead of word-level likelihood.
1 code implementation • 12 Sep 2018 • Xiaoyuan Yi, Maosong Sun, Ruoyu Li, Zonghan Yang
Different from previous methods, our model explicitly maintains topics and informative limited history in a neural memory.
no code implementations • CONLL 2018 • Xiaoyuan Yi, Ruoyu Li, Maosong Sun
As a precious part of the human cultural heritage, Chinese poetry has influenced people for generations.
no code implementations • 6 Apr 2016 • Xiaoyuan Yi, Ruoyu Li, Maosong Sun
We take the generation of Chinese classical poem lines as a sequence-to-sequence learning problem, and build a novel system based on the RNN Encoder-Decoder structure to generate quatrains (Jueju in Chinese), with a topic word as input.