Scholar
Xinting Huang
Google Scholar ID: QmyPDWQAAAAJ
Tencent
Citations & Impact
Citations
2,469
H-index
13
i10-index
14
Publications
20
Co-authors
15
list available
Publications
Resume (English only)
Academic Achievements
- Published multiple papers in the areas of preference alignment and hallucination mitigation, efficient training, and synthetic data, including:
- - 'Knowledge Verification to Nip Hallucination in the Bud' (EMNLP, 2024)
- - 'SEGO: Sequential Subgoal Optimization for Mathematical Problem-Solving' (ACL, 2024)
- - 'BBA: Bi-Modal Behavioral Alignment for Reasoning with Large Vision-Language Models' (ACL Findings, 2024)
- - 'Explore-Instruct: Enhancing Domain-Specific Instruction Coverage through Active Exploration' (EMNLP, 2023)
- - 'Siren's Song in the AI Ocean: A Survey on Hallucination in Large Language Models' (Preprint, 2023)
- - 'Knowledge Fusion of Large Language Models' (ICLR, 2024)
- - 'Pre-training Multi-party Dialogue Models with Latent Discourse Inference' (ACL, 2023)
- - 'TeGit: Generating High-Quality Instruction-Tuning Data with Text-Grounded Task Design' (NAACL, 2024)
- - 'FuseChat: Knowledge Fusion of Chat Models' (Technical Report, 2024)
Research Experience
- Interned at Microsoft Research Asia and Bytedance AI Lab. Currently a Senior Researcher at Tencent, working on LLMs alignment.
Education
- PhD from The University of Melbourne in 2022, supervised by Prof. Rui Zhang; undergraduate from Renmin University of China.
Background
- A senior researcher at Tencent, focusing on large language model (LLM) alignment. Research interests include preference alignment, hallucination mitigation, efficient training, and synthetic data.
Miscellany
- Contact: timhuangxt AT gmail.com, Tel: 18664590062.
