BEng, PhD
Lecturer
- About
-
- Email Address
- ruizhe.li@abdn.ac.uk
- School/Department
- School of Natural and Computing Sciences
Biography
Ruizhe Li is a Lecturer (Assistant Professor) at the Aberdeen - South China Normal University (SCNU) Joint Institute in the School of Natural Computing Sciences/Business School, the University of Aberdeen. He is associated with Natural Language Processing (NLP) group, the Machine Learning (ML) group and Edinburgh clinical NLP group. He was a postdoc research fellow in the Web Intelligence Group, affiliated with the Centre for Artificial Intelligence at University College London (UCL). He received his PhD from the Sheffield NLP group, the Department of Computer Science at the University of Sheffield. He completed his bachelor's degree in BEng Electronic Information & Engineering at Shanghai University, China.
Research Interests: Interpretability of LLMs, Multimodal LLMs
For more details, please visit my personal website: https://www.ruizhe.space/
I am looking for highly motivated PhD students interested in NLP and ML. Please contact me for any enquiries.
Latest Publications
Anchored Answers: Unravelling Positional Bias in GPT-2's Multiple-Choice Questions
Working Papers: Preprint Papers- [ONLINE] file:///C:/Users/s01gv9/Downloads/2405.03205v2.pdf
- [ONLINE] DOI: https://doi.org/10.48550/arXiv.2405.03205
Listen Again and Choose the Right Answer: A New Paradigm for Automatic Speech Recognition with Large Language Models
Working Papers: Preprint Papers- [ONLINE] DOI: https://doi.org/10.48550/arXiv.2405.10025
Overcoming Catastrophic Forgetting by Exemplar Selection in Task-oriented Dialogue System
Working Papers: Preprint Papers- [ONLINE] file:///C:/Users/s01gv9/Downloads/2405.10992v1.pdf
- [ONLINE] DOI: https://doi.org/10.48550/arXiv.2405.1099
Large Language Models are Efficient Learners of Noise-Robust Speech Recognition
Chapters in Books, Reports and Conference Proceedings: Conference ProceedingsIt's Never Too Late: Fusing Acoustic Information into Large Language Models for Automatic Speech Recognition
Chapters in Books, Reports and Conference Proceedings: Conference Proceedings
Prizes and Awards
- Best Poster Award at the 2nd Workshop on Multimodal AI 2024
- ACL 2023 Area Chair Award
- Outstanding Reviewer at EMNLP 2020
- Best Paper Runner-up at INLG 2019
- Research
-
Research Overview
- Multimodality LLMs
- Knowledge Graphs
- LLMs Reasoning
- Explainable LLMs.
Research Areas
Accepting PhDs
I am currently accepting PhDs in Computing Science.
Please get in touch if you would like to discuss your research ideas further.
Research Specialisms
- Artificial Intelligence
- Natural Language Processing
- Machine Learning
Our research specialisms are based on the Higher Education Classification of Subjects (HECoS) which is HESA open data, published under the Creative Commons Attribution 4.0 International licence.
Funding and Grants
PI, Deep-learning-based Modelling of Electromechanical Coupling for Bone Conduction Transducers, The Royal Society International Exchanges 2023 Cost Share (NSFC), £24,000, 2024-2026
- Teaching
-
- Publications
-
Page 1 of 1 Results 1 to 24 of 24
Anchored Answers: Unravelling Positional Bias in GPT-2's Multiple-Choice Questions
Working Papers: Preprint Papers- [ONLINE] file:///C:/Users/s01gv9/Downloads/2405.03205v2.pdf
- [ONLINE] DOI: https://doi.org/10.48550/arXiv.2405.03205
Listen Again and Choose the Right Answer: A New Paradigm for Automatic Speech Recognition with Large Language Models
Working Papers: Preprint Papers- [ONLINE] DOI: https://doi.org/10.48550/arXiv.2405.10025
Overcoming Catastrophic Forgetting by Exemplar Selection in Task-oriented Dialogue System
Working Papers: Preprint Papers- [ONLINE] file:///C:/Users/s01gv9/Downloads/2405.10992v1.pdf
- [ONLINE] DOI: https://doi.org/10.48550/arXiv.2405.1099
Large Language Models are Efficient Learners of Noise-Robust Speech Recognition
Chapters in Books, Reports and Conference Proceedings: Conference ProceedingsIt's Never Too Late: Fusing Acoustic Information into Large Language Models for Automatic Speech Recognition
Chapters in Books, Reports and Conference Proceedings: Conference ProceedingsSimulated Task Oriented Dialogues for Developing Versatile Conversational Agents
Chapters in Books, Reports and Conference Proceedings: Conference ProceedingsGenTranslate: Large Language Models are Generative Multilingual Speech and Machine Translators
Working Papers: Preprint Papers- [ONLINE] DOI: https://doi.org/10.48550/arXiv.2402.06894
- [OPEN ACCESS] http://aura.abdn.ac.uk/bitstream/2164/22841/1/2402.06894v1.pdf
Leveraging A Medical Knowledge Graph into Large Language Models for Diagnosis Prediction
Working Papers: Preprint Papers- [ONLINE] http://arxiv.org/pdf/2308.14321v1
- [ONLINE] DOI: https://doi.org/10.48550/arXiv.2308.14321
Cross-Modal Global Interaction and Local Alignment for Audio-Visual Speech Recognition
Chapters in Books, Reports and Conference Proceedings: Conference Proceedings- [ONLINE] DOI: https://doi.org/10.24963/ijcai.2023/564
- [ONLINE] View publication in Scopus
Noise-aware Speech Enhancement using Diffusion Probabilistic Model
Working Papers: Preprint Papers- [ONLINE] http://arxiv.org/abs/2307.08029v1
- [ONLINE] DOI: https://doi.org/10.48550/arXiv.2307.08029
- [OPEN ACCESS] http://aura.abdn.ac.uk/bitstream/2164/21469/1/2307.08029v1.pdf
Hearing Lips in Noise: Universal Viseme-Phoneme Mapping and Transfer for Robust Audio-Visual Speech Recognition
Chapters in Books, Reports and Conference Proceedings: Conference ProceedingsMIR-GAN: Refining Frame-Level Modality-Invariant Representations with Adversarial Network for Audio-Visual Speech Recognition
Chapters in Books, Reports and Conference Proceedings: Conference ProceedingsGradient Remedy for Multi-Task Learning in End-to-End Noise-Robust Speech Recognition
Chapters in Books, Reports and Conference Proceedings: Conference Proceedings- [ONLINE] DOI: https://doi.org/10.1109/ICASSP49357.2023.10096615
Cross-Modal Global Interaction and Local Alignment for Audio-Visual Speech Recognition
Working Papers: Preprint PapersOn the Latent Holes of VAEs for Text Generation
Contributions to Conferences: Oral Presentations- [ONLINE] DOI: https://doi.org/10.48550/arXiv.2110.03318
- [OPEN ACCESS] http://aura.abdn.ac.uk/bitstream/2164/18658/1/2110.03318v1.pdf
On the Low-density Latent Regions of VAE-based Language Models
Chapters in Books, Reports and Conference Proceedings: Conference ProceedingsAffective Decoding for Empathetic Response Generation
Chapters in Books, Reports and Conference Proceedings: Conference Proceedings- [ONLINE] View publication in Scopus
Improving Variational Autoencoder for Text Modelling with Timestep-Wise Regularisation
Contributions to Conferences: Oral PresentationsLatent Space Factorisation and Manipulation via Matrix Subspace Projection
Contributions to Journals: Conference ArticlesDGST: A dual-generator network for text style transfer
Chapters in Books, Reports and Conference Proceedings: Conference Proceedings- [ONLINE] View publication in Scopus
Latent Space Factorisation and Manipulation via Matrix Subspace Projection
Chapters in Books, Reports and Conference Proceedings: Conference Proceedings- [ONLINE] http://arxiv.org/abs/1907.12385v3
- [OPEN ACCESS] http://aura.abdn.ac.uk/bitstream/2164/18942/1/1907.12385v3.pdf
- [ONLINE] View publication in Scopus
A Stable Variational Autoencoder for Text Modelling
Contributions to Conferences: Oral PresentationsA Dual-Attention Hierarchical Recurrent Neural Network for Dialogue Act Classification
Chapters in Books, Reports and Conference Proceedings: Conference Proceedings- [ONLINE] https://www.aclweb.org/anthology/K19-1036.pdf
- [ONLINE] DOI: https://doi.org/10.18653/v1/K19-1036
- [OPEN ACCESS] http://aura.abdn.ac.uk/bitstream/2164/13116/2/1810.09154v3.pdf
- [OPEN ACCESS] http://aura.abdn.ac.uk/bitstream/2164/13116/3/Li_etal_Dual_Attention_VOR.pdf
- [ONLINE] https://www.aclweb.org/anthology/K19-1/
ABDN at SemEval-2018 Task 10: Recognising Discriminative Attributes using Context Embeddings and WordNet
Chapters in Books, Reports and Conference Proceedings: Conference Proceedings- [ONLINE] http://www.aclweb.org/anthology/S18-1169
- [ONLINE] View publication in Scopus