Jaeri Lee

Jongjin Kim, Hyunsik Jeon, Jaeri Lee et al.

When recommending personalized top-$k$ items to users, how can we recommend the items diversely to them while satisfying their needs? Aggregately diversified recommender systems aim to recommend a variety of items across whole users without sacrificing the recommendation accuracy. They increase the exposure opportunities of various items, which in turn increase potential revenue of sellers as well as user satisfaction. However, it is challenging to tackle aggregate-level diversity with a matrix factorization (MF), one of the most common recommendation model, since skewed real world data lead to skewed recommendation results of MF. In this work, we propose DivMF (Diversely Regularized Matrix Factorization), a novel matrix factorization method for aggregately diversified recommendation. DivMF regularizes a score matrix of an MF model to maximize coverage and entropy of top-$k$ recommendation lists to aggregately diversify the recommendation results. We also propose an unmasking mechanism and carefully designed mi i-batch learning technique for accurate and efficient training. Extensive experiments on real-world datasets show that DivMF achieves the state-of-the-art performance in aggregately diversified recommendation.

Minjun Kim, Jaeri Lee, Jongjin Kim et al.

How can we accurately quantize a pre-trained Vision Transformer model? Quantization algorithms compress Vision Transformers (ViTs) into low-bit formats, reducing memory and computation demands with minimal accuracy degradation. However, existing methods rely on uniform precision, ignoring the diverse sensitivity of ViT components to quantization. Metric-based Mixed Precision Quantization (MPQ) is a promising alternative, but previous MPQ methods for ViTs suffer from three major limitations: 1) coarse granularity, 2) mismatch in metric scale across component types, and 3) quantization-unaware bit allocation. In this paper, we propose LampQ (Layer-wise Mixed Precision Quantization for Vision Transformers), an accurate metric-based MPQ method for ViTs to overcome these limitations. LampQ performs layer-wise quantization to achieve both fine-grained control and efficient acceleration, incorporating a type-aware Fisher-based metric to measure sensitivity. Then, LampQ assigns bit-widths optimally through integer linear programming and further updates them iteratively. Extensive experiments show that LampQ provides the state-of-the-art performance in quantizing ViTs pre-trained on various tasks such as image classification, object detection, and zero-shot quantization.