Shengqiong Wu

I am currently a second-year Ph.D. student at NExT++ Research Center, advised by Prof. Tat-Seng Chua in School of Computing at National University of Singapore, after I obtained my M.S. and Bachelor degrees from Wuhan University.

My research interest mainly lies in Scene Graph-based Vision-Language Understanding, with the contexts of Multimodal Large Language Models and Diffusion Model, etc. Previously, I focused on syntax-aided NLP, i.e., employing external syntactic and linguistic representation to aid NLP tasks.

Email  /  Github  /  Google Scholar

• Shengqiong Wu, Hao Fei, Xiangtai Li, Jiayi Ji, Hanwang Zhang, Tat-Seng Chua, Shuicheng Yan.  Towards Semantic Equivalence of Tokenization in Multimodal LLM. arXiv. 2024.  [Code][pdf]
• Meng Luo, Hao Fei*, Bobo Li, Shengqiong Wu, Qian Liu, Soujanya Poria, Erik Cambria, Mong-Li Lee, Wynne Hsu.  PanoSent: A Panoptic Sextuple Extraction Benchmark for Multimodal Conversational Aspect-based Sentiment Analysis. In Proceedings of ACM MM. 2024.   (Oral).
• Shengqiong Wu, Hao Fei, Leigang Qu, Wei Ji, Tat-Seng Chua.  NExT-GPT: Any-to-Any Multimodal Large Language Model. In Proceedings of ICML. 2024.   (Oral) [Code | 3.2k 🌟][pdf]
• Hao Fei, Shengqiong Wu, Wei Ji, Hanwang Zhang, Meishan Zhang, Mong-Li Lee, Wynne Hsu.  Video-of-Thought: Step-by-Step Video Reasoning from Perception to Cognition. In Proceedings of ICML. 2024.   (Oral) [Code][pdf]
• Hao Fei, Shengqiong Wu, Wei Ji, Hanwang Zhang, Tat-Seng Chua.  Dysen-VDM: Empowering Dynamics-aware Text-to-Video Diffusion with LLMs. In Proceedings of CVPR. 2024.  [Code][pdf]
• Hao Fei, Shengqiong Wu, Hanwang Zhang, Tat-Seng Chua, Shuicheng Yan.  VITRON: A Unified Pixel-level Vision LLM for Understanding, Generating, Segmenting, Editing. arXiv. 2024.  [Code][pdf]

• Shengqiong Wu, Hao Fei, Hanwang Zhang, Tat-Seng Chua.  Imagine That! Abstract-to-Intricate Text-to-Image Synthesis with Scene Graph Hallucination Diffusion. In Proceedings of NeurIPS. 2023.  (long, poster)  [Code][pdf]
• Leigang Qu*, Shengqiong Wu*, Hao Fei, Liqiang Nie, Tat-Seng Chua.  LayoutLLM-T2I: Eliciting Layout Guidance from LLM for Text-to-Image Generation. In Proceedings of ACM MM. 2023.  (*: equal contribution, long)  [Code][pdf]
• Bobo Li, Hao Fei, Yuhan Wu, Jinsong Zhang, Shengqiong Wu, Jingye Li, Yijiang Liu, Lizi Liao, Tat-Seng Chua, Fei Li, Donghong Ji.  DiaASQ: A benchmark of conversational aspect-based sentiment quadruple analysis.In Proceedings of ACL. 2023.  (long, poster)  [Code][pdf]
• Shengqiong Wu, Hao Fei, Yixin Cao, Lidong Bing, Tat-Seng Chua.  Information Screening whilst Exploiting! Multimodal Relation Extraction with Feature Denoising and Multimodal Topic Modeling. In Proceedings of ACL. 2023.  (long, poster, paper award nomination, 1.6%)  [Code][pdf]
• Shengqiong Wu, Hao Fei, Wei Ji, Tat-Seng Chua.  Cross2StrA: Unpaired Cross-lingual Image Captioning with Cross-lingual Cross-modal Structure-pivoted Alignment. In Proceedings of ACL. 2023.  (long, oral)  [pdf]

• Hao Fei, Shengqiong Wu, Jingye Li, Bobo Li, Fei Li, Libo Qin, Meishan Zhang, Min Zhang, Tat-Seng Chua.  LasUIE: Unifying information extraction with latent adaptive structure-aware generative language model. In Proceedings of NeurIPS. 2022.  (long, poster) [Code][pdf]
• Shunjie Chen, Xiaochuan Shi, Jingye Li, Shengqiong Wu, Hao Fei, Fei Li, Donghong Ji.  Joint Alignment of Multi-Task Feature and Label Spaces for Emotion Cause Pair Extraction. In Proceedings of COLING. 2022.  (long, poster) [Code][pdf]
• Hu Cao, Jingye Li, Fangfang Su, Fei Li, Hao Fei, Shengqiong Wu, Bobo Li, Liang Zhao and Donghong Ji.  OneEE: A One-Stage Framework for Fast Overlapping and Nested Event Extraction. In Proceedings of COLING. 2022.  (long, oral) [Code][pdf]
• Shengqiong Wu, Hao Fei, Fei Li, Meishan Zhang, Yijiang Liu, Chong Teng, Donghong Ji.  Mastering the Explicit Opinion-Role Interaction: Syntax-Aided Neural Transition System for Unified Opinion Role Labeling. In Proceedings of AAAI. 2022.  (long, online) [Code][pdf]
• Jingye Li, Hao Fei, Jiang Liu, Shengqiong Wu, Meishan Zhang, Chong Teng, Donghong Ji, Fei Li.  Unified named entity recognition as word-word relation classification. In Proceedings of AAAI. 2022.  (long, online) [Code][pdf]

• Shengqiong Wu, Hao Fei, Yafeng Ren, Donghong Ji, Jingye Li.  Learn from Syntax: Improving Pair-wise Aspect and Opinion Terms Extraction with Rich Syntactic Knowledge. In Proceedings of IJCAI. 2021.  (long, online) [Code][pdf]
• Shengqiong Wu, Hao Fei, Yafeng Ren, Bobo Li, Fei Li and Donghong Ji.   High-order Pair-wise Aspectand Opinion Terms Extraction with Edge-enhanced Syntactic Graph Convolution. IEEE/ACM TASLP. 2021. [Code][pdf]
• Shengqiong Wu, Bobo Li, Dongdong Xie, Chong Teng and Donghong Ji.  Neural Transition Model for Aspect-based Sentiment Triplet Extraction with Triplet Memory. Neurocomputing. 2021.   [pdf]

• ShengqiongWu, Hao Fei and Donghong Ji.  Aggressive Language Detection with Joint Text Normalization via Adversarial Multi-task Learning. In Proceedings of NLPCC. 2020.  (long, oral) [Code][pdf]

1. [2024/01]       The Path to AGI: Achieving Modality Unification with NExT-GPT. Qingyuan Talk.
2. [2023/12]       NExT-GPT: Any-to-Any Multimodal LLM. AI New Youth.
3. [2022/12]       Deep Learning based Natural Language Processing: A Survey and Outlook. Jianghan University, China.
4. [2021/10]       Comparison of Aspect-based Sentiment Analysis based on span model and transition model. Wuhan University, China.
5. [2021/05]       Learn from Syntax: Improving Pair-wise Aspect and Opinion Terms Extraction with Rich Syntactic Knowledge. CIPS, Youth Working Committee.