Chenyang Li

Also published as: 晨阳


2024

pdf bib
基于参数高效微调与半监督学习的空间语义理解
Chenyang Li (李晨阳) | Long Zhang (张龙) | Qiusheng Zheng (郑秋生)
Proceedings of the 23rd Chinese National Conference on Computational Linguistics (Volume 3: Evaluations)

“本文介绍了我们在第二十三届中文计算语言大会的第四届中文空间语义理解评测任务中提交的参赛模型。该任务旨在测试机器的中文语义理解水平。现有研究显示,机器的中文语义理解水平与人类平均水平相比仍有较大差距。近年来,生成式大规模语言模型在自然语言处理任务中展现了出色的生成和泛化能力。在本次评测中,我们采用了对Qwen1.5-7b模型进行高效微调的方法,以端到端的形式实现空间语义的推理过程,并结合prompt优化和半监督学习提升推理表现。实验结果表明,我们的模型在该任务中取得了领先的效果。”

pdf bib
基于深度学习模型的中小学作文修辞识别与理解评测
Chenyang Li (李晨阳) | Long Zhang (张龙) | Qiusheng Zheng (郑秋生)
Proceedings of the 23rd Chinese National Conference on Computational Linguistics (Volume 3: Evaluations)

“在中小学生的学习进程中,修辞手法是阅读和写作技巧的核心,也是优秀文学作品的关键元素。然而,识别与理解学生文章中的修辞使用需要大量的人工,为教师的作文评估和教学提出了挑战。最近的研究开始使用计算机技术来自动评审作文,其中修辞的使用是评估的重要部分。本文介绍了我们在第二十三届中文计算语言大会中中小学作文修辞识别与理解评测中的所用的参赛方法。在本次评测中,我们针对不同任务,分别使用了传统模型分类模型和大模型,再利用伪标签、数据增强等方法提升模型性能。实验结果表明,我们的方法取得了较为先进的效果。”

2023

pdf bib
CCL23-Eval 任务6系统报告:基于深度学习的电信网络诈骗案件分类(System Report for CCL23-Eval Task 6: Classification of Telecom Internet Fraud Cases Based on Deep Learning)
Chenyang Li (李晨阳) | Long Zhang (张龙) | Zhongjie Zhao (赵中杰) | Hui Guo (郭辉)
Proceedings of the 22nd Chinese National Conference on Computational Linguistics (Volume 3: Evaluations)

“文本分类任务作为自然语言处理领域的基础任务,在面向电信网络诈骗领域的案件分类中扮演着至关重要的角色,对于智能化案件分析具有重大意义和深远影响。本任务的目的是对给定案件描述文本进行分类,案件文本包含对案件的经过脱敏处理后的整体描述。我们首先采用Ernie预训练模型对案件内容进行微调的方法得到每个案件的类别,再使用伪标签和模型融合方法对目前的F1值进行提升,最终在CCL23-Eval任务6电信网络诈骗案件分类评测中取得第二名的成绩,该任务的评价指标F1值为0.8628,达到了较为先进的检测效果。”

pdf bib
WordArt Designer: User-Driven Artistic Typography Synthesis using Large Language Models
Jun-Yan He | Zhi-Qi Cheng | Chenyang Li | Jingdong Sun | Wangmeng Xiang | Xianhui Lin | Xiaoyang Kang | Zengke Jin | Yusen Hu | Bin Luo | Yifeng Geng | Xuansong Xie
Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing: Industry Track

This paper introduces WordArt Designer, a user-driven framework for artistic typography synthesis, relying on the Large Language Model (LLM). The system incorporates four key modules: the LLM Engine, SemTypo, StyTypo, and TexTypo modules. 1) The LLM Engine, empowered by the LLM (e.g. GPT-3.5), interprets user inputs and generates actionable prompts for the other modules, thereby transforming abstract concepts into tangible designs. 2) The SemTypo module optimizes font designs using semantic concepts, striking a balance between artistic transformation and readability. 3) Building on the semantic layout provided by the SemTypo module, the StyTypo module creates smooth, refined images. 4) The TexTypo module further enhances the design’s aesthetics through texture rendering, enabling the generation of inventive textured fonts. Notably, WordArt Designer highlights the fusion of generative AI with artistic typography. Experience its capabilities on ModelScope: https://www.modelscope.cn/studios/WordArt/WordArt.

2021

pdf bib
Pretrain-Finetune Based Training of Task-Oriented Dialogue Systems in a Real-World Setting
Manisha Srivastava | Yichao Lu | Riley Peschon | Chenyang Li
Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies: Industry Papers

One main challenge in building task-oriented dialogue systems is the limited amount of supervised training data available. In this work, we present a method for training retrieval-based dialogue systems using a small amount of high-quality, annotated data and a larger, unlabeled dataset. We show that pretraining using unlabeled data can bring better model performance with a 31% boost in Recall@1 compared with no pretraining. The proposed finetuning technique based on a small amount of high-quality, annotated data resulted in 26% offline and 33% online performance improvement in Recall@1 over the pretrained model. The model is deployed in an agent-support application and evaluated on live customer service contacts, providing additional insights into the real-world implications compared with most other publications in the domain often using asynchronous transcripts (e.g. Reddit data). The high performance of 74% Recall@1 shown in the customer service example demonstrates the effectiveness of this pretrain-finetune approach in dealing with the limited supervised data challenge.