
Bin Li
Hunan University
pre-trained language model
knowledge graph
self-supervised learning
multi-task
multimodal learning
video understanding
knowledge transfer
question-answering
prompt tuning
two-staged cross-modal fusion
medical instructional video classification
prompt-based tuning
personality and interpersonal reactivity prediction
continuing pre-training
emotional classification
7
presentations
4
number of views
SHORT BIO
I am a direct doctoral student from the School of Electrical and Information Engineering, Hunan University, majoring in Control Science and Engineering. My supervisor is Professor Li Shutao. My main research interests include dialogue generation, human-computer interaction, natural language processing, etc. He has published 4 foreign conference papers, published 1 patent and 2 soft-authored papers. The designed and developed "Q&A Assistant for Gastroenterology" was demonstrated at the 2021 National Conference on Knowledge Graph and Semantic Computing (CCKS2021). In terms of projects, I have successively participated in the national key R&D project "Multimodal Integration of Robotic Natural Interaction"; the school's key project "Research on Human-Computer Interaction Key Technologies of Intelligent Connected Vehicles"; the National Natural Science Foundation of China's general project "Services" "Multimodal Natural Interaction Understanding and Feedback of Robots", assisted in the development of the intelligent guidance robot "Xiaolu" for the medical field and a multimodal integrated human-machine natural interaction system.
Presentations

MedConQA: Medical Conversational Question Answering System based on Knowledge Graphs
Fei Xia and 2 other authors

Knowledge Transfer with Visual Prompt in Multi-modal Dialogue Understanding and Generation
MinJun Zhu and 2 other authors

LingJing at SemEval-2022 Task 1: Multi-task Self-supervised Pre-training for Multilingual Reverse Dictionary
Bin Li

Applying DeBERTa to Lexical-level Presupposed Relation Taxonomy with Knowledge Transfer
Bin Li

VPAI Lab at MedVidQA 2022: A Two Stage Cross-modal Fusion Method for Medical Instructional Video Classification
Bin Li

Continuing Pre-trained Model with Multiple Training Strategies for Emotional Classification
Bin Li

Prompt-based Pre-trained Model for Personality and Interpersonal Reactivity Prediction
Bin Li