Evaluation Schedule
Invited Talk 1
Speaker:Jing Shao
Title:Building Security Evaluation Systems for Large Models and Intelligent Agents: Near-term Vulnerability Discovery and Long-term Risk Prediction
Biography:Jing Shao received her Ph.D. from MMLab at The Chinese University of Hong Kong. She previously served as Research Director at SenseTime Technology and is currently the Co-Director of the Safe and Trustworthy AI Center at Shanghai AI Laboratory and a Young Scientist, also serving as a doctoral supervisor at Shanghai Jiao Tong University and Fudan University. Her research focuses on autonomous, controllable, and trustworthy AI, with particular emphasis on security evaluation and value alignment of multimodal large models and intelligent agents. She has published over 50 papers in top international conferences and journals with more than 11,100 citations, and received the Outstanding Paper Award at ACL2024. She has been selected for Shanghai Oriental Talent Program and Shenzhen Overseas High-level Talent Program, is a member of the China Association for Science and Technology Workers, and was listed in the Stanford Top 2% Scientists list for 2024. She holds over 20 authorized patents and has validated applications in smart cities, intelligent transportation, and smartphones, receiving the "Shenzhen AI Technology Progress Award" in 2022.
Invited Talk 2
Speaker:Yue Zhang
Title:Comprehensive Robust Evaluation of Large Model Reasoning Capabilities
Biography:Yue Zhang is a Professor at Westlake University and Associate Dean of the School of Engineering. He received his B.S. in Computer Science from Tsinghua University in 2003, Ph.D. in Computer Science from Oxford University in 2009, and conducted postdoctoral research at Cambridge University from 2010-2012. His main research areas include natural language processing, language models, and trustworthy artificial intelligence. He is the author of "Natural Language Processing: A Machine Learning Perspective" published by Cambridge University Press and co-editor of Oxford Bibliography in Natural Language Processing. He has served as Program Committee Chair for top international conferences including CCL 2020 and EMNLP 2022, and as an editorial board member for journals such as TACL and TASLP. He has received best paper awards at ACL 2018 (nomination), COLING 2018, IALP 2017, SemEval 2020, and ACL 2023 (nomination). He has been selected for the Stanford Global Top 2% Scientists list and Elsevier China Highly Cited Researchers list.
Evaluation Schedule
Time:August 12
Venue:TBD
| Time Slot | Duration (Minutes) | Program |
|---|---|---|
| 8:30-8:35 | 5 | Opening Ceremony |
| 8:35-9:20 | 45 | Invited Talk 1: Shao Jing Building Security Evaluation Systems for Large Models and Intelligent Agents |
| 9:20-9:50 | 30 | Task 1: The 5th Spatial Semantic Understanding Evaluation |
| 9:50-10:10 | 20 | Task 2: The 3rd Chinese Frame Semantic Parsing Evaluation |
| 10:10-10:30 | 20 | Task 3: The 5th Chinese Abstract Meaning Representation Parsing Evaluation |
| 10:30-10:40 | 10 | Break |
| 10:40-11:10 | 30 | Task 4: The 1st Chinese Factual Reasoning Evaluation |
| 11:10-11:40 | 30 | Task 5: The 1st Chinese Poetry Appreciation Evaluation |
| 11:40-12:10 | 30 | Task 6: The 2nd Chinese Essay Rhetoric Recognition and Understanding |
| 12:10-13:40 | Lunch | |
| 13:40-14:25 | 45 | Invited Talk 2: Zhang Yue Comprehensive Robust Evaluation of Large Model Reasoning Capabilities |
| 14:25-14:55 | 30 | Task 7: The 1st Chinese Literary Language Understanding Evaluation (Debate) |
| 14:55-15:10 | 15 | Task 8: Chinese Electronic Medical Record ICD Diagnostic Coding Evaluation |
| 15:10-15:40 | 30 | Task 9: Traditional Chinese Medicine Syndrome Differentiation and Herbal Prescription Generation Evaluation |
| 15:40-15:50 | 10 | Break |
| 15:50-16:20 | 30 | Task 10: Fine-grained Chinese Hate Speech Recognition Evaluation |
| 16:20-16:50 | 30 | Task 11: College Students' Chinese Character Handwriting Quality Evaluation |
| 16:50-17:20 | 30 | Task 12: Entity Relation Triple Extraction Evaluation for Chinese Speech |
| 17:20-18:00 | 40 | Panel Discussion |