6 papers
FIBER: A Differentially Private Optimizer with Filter-Aware Innovation Bias Correction
Duc Dm, Thao Do, Minh Son Hoang +3
Differentially private (DP) training protects individual examples by adding noise to gradients, but the injected noise interacts nontrivially with adaptive optimizers. Recent DP me…
SAT: Selective Aggregation Transformer for Image Super-Resolution
Dinh Phu Tran, Thao Do, Saad Wazir +3
Transformer-based approaches have revolutionized image super-resolution by modeling long-range dependencies. However, the quadratic computational complexity of vanilla self-attenti…
LooComp: Leverage Leave-One-Out Strategy to Encoder-only Transformer for Efficient Query-aware Context Compression
Thao Do, Dinh Phu Tran, An Vo +2
Efficient context compression is crucial for improving the accuracy and scalability of question answering. For the efficiency of Retrieval Augmented Generation, context should be d…
Knowing When to Answer: Adaptive Confidence Refinement for Reliable Audio-Visual Question Answering
Dinh Phu Tran, Jihoon Jeong, Saad Wazir +4
We present a formal problem formulation for \textit{Reliable} Audio-Visual Question Answering (-AVQA), where we prefer abstention over answering incorrectly. While rec…
VDInstruct: Zero-Shot Key Information Extraction via Content-Aware Vision Tokenization
Son Nguyen, Giang Nguyen, Hung Dao +2
Key Information Extraction (KIE) underpins the understanding of visual documents (e.g., receipts and contracts) by extracting precise semantic content and accurately capturing spat…
Reference-Based Post-OCR Processing with LLM for Precise Diacritic Text in Historical Document Recognition
Thao Do, Dinh Phu Tran, An Vo +1
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 s…