Xudong Hong

Research Interests

My research interests span across computer science, computational linguistics, and cognitive science. I use natural language processing and deep learning methods to investigate questions including:

My current focus is to obtain prototypical event representations with inner-event structures and intra-event relationships using multimodal learning, in order to generate semantically coherent and pragmatically informative stories.

In Submission

A paper on data efficiency.


Xudong Hong, Margarita Ryzhova, Daniel Adrian Biondi, Vera Demberg. Do large language models and humans have similar behaviors in causal inference with script knowledge? (arxiv 2023)

Xudong Hong, Sharid Loáiciga, Asad Sayeed. A surprisal oracle for active curriculum language modeling (BabyLM@CoNLL 2023)

Nina Shvetsova, Anna Kukleva, Xudong Hong, Christian Rupprecht, Bernt Schiele, Hilde Kuehne. Howtocaption: Prompting llms to transform video annotations at scale (arxiv 2023)

Xudong Hong, Khushboo Mehra, Asad Sayeed, Vera Demberg, Bernt Schiele. Visually Grounded Story Generation Challenge (INLG 2023)

Xudong Hong, Vera Demberg, Asad Sayeed, Qiankun Zheng, Bernt Schiele. Visual Coherence Loss for Coherent and Visually Grounded Story Generation Findings of ACL 2023.

Xudong Hong, Asad Sayeed, Khushboo Mehra, Vera Demberg, and Bernt Schiele. Visual Writing Prompts: Character-Grounded Story Generation with Curated Image Sequences. Transactions of the Association for Computational Linguistics (TACL), 11, 2023.

Dongqi Pu*, Xudong Hong*, Pin-Jie Lin*, Ernie Chang, and Vera Demberg. Two-Stage Movie Script Summarization: An Efficient Method For Low-Resource Long Document Summarization. In Proceedings of The Workshop on Automatic Summarization for Creative Writing (Creativesumm@COLING), pages 57–66, October 2022.

Han Cao, Xudong Hong, Heike Tost, Andreas Meyer-Lindenberg, and Emanuel Schwarz. Advancing translational research in neuroscience through multi-task learning. Frontiers in Psychiatry, 13, 2022.

Xudong Hong, Rakshith Shetty, Asad Sayeed, Khushboo Mehra, Vera Demberg, Bernt Schiele. Diverse and Relevant Visual Storytelling with Scene Graph Embeddings. In Proceedings of the 24th Conference on Computational Natural Language Learning (CoNLL), 2020.

Xudong Hong, Ernie Chang, and Vera Demberg. Improving language generation from feature-rich tree-structured data with relational graph convolutional encoders. In Proceedings of the 2nd Workshop on Multilingual Surface Realisation (MSR@EMNLP), 2019.

Xudong Hong, Asad Sayeed, and Vera Demberg. Learning distributed event representations with a multi-task approach. In Proceedings of the Seventh Joint Conference on Lexical and Computational Semantics (*SEM), pages 11–21, 2018.

Asad Sayeed, Xudong Hong, and Vera Demberg. Roleo: visualising thematic fit spaces on the web. In Proceedings of ACL-2016 System Demonstrations (ACL demo), pages 139–144, 2016.

Please find the full list of my publications at Google Scholar

Contact Information

Main email: x[lastname]@coli.uni-saarland.de

Other emails: x[lastname]@mpi-inf.mpg.de / tony.xudong.hong@gmail.com

Phone (mobile): +49 174 733 1336

Phone (fixed): +49 681 302 70020

Address: Campus C72 2.02, 66123 Saarbrücken