Tianshu Zhang

Tianshu Zhang's recent work focuses on making large language models and vision-language models more trustworthy, particularly by reducing 'hallucinations' (cases where AI systems generate false or unsupported information). Related work spans few-shot learning (teaching models new tasks with very little data), federated learning (training models across decentralized data sources without sharing raw data), and text-to-SQL systems (translating natural language into database queries). Earlier publications also touch on biometric recognition, human body/pose estimation from images, and some unrelated topics in number theory and genetic algorithms.

Publication output was low and sporadic from 2017–2019, increased notably in 2020 and especially 2023, and has continued at a moderate, steady pace through 2024–2026.

Generated by claude-sonnet-5 from public bibliographic data · Jul 20, 2026