Thao Do — accepted AI-conference papers

2026

FIBER: A Differentially Private Optimizer with Filter-Aware Innovation Bias Correction

MINH DUC DO, Thao Do, Minh Hoang, Anh Le Duc Tran, Daeyoung Kim, Huy L Nguyen

ICML 2026poster

Differentially private (DP) training protects individual examples by adding noise to gradients, but the injected noise interacts nontrivially with adaptive optimizers. Recent DP methods temporally filter privatized gradients to reduce variance; however, filtering also changes the DP noise statistics…

Cited by 0 Source ↗Scholar ↗

2025

Reference-Based Post-OCR Processing with LLM for Precise Diacritic Text in Historical Document Recognition

Thao Do, Dinh Phu Tran, An Vo, Daeyoung Kim

AAAI 2025technical

Extracting fine-grained OCR text from aged documents in diacritic languages remains challenging due to unexpected artifacts, time-induced degradation, and lack of datasets. While standalone spell correction approaches have been proposed, they show limited performance for historical documents due to…

Cited by 0 Source ↗PDF ↗code ↗Scholar ↗