Search Results for author: Minjae Lee

Found 14 papers, 3 papers with code

LoL-PIM: Long-Context LLM Decoding with Scalable DRAM-PIM System

no code implementations28 Dec 2024 Hyucksung Kwon, Kyungmo Koo, Janghyeon Kim, Woongkyu Lee, Minjae Lee, Hyungdeok Lee, Yousub Jung, JaeHan Park, Yosub Song, Byeongsu Yang, Haerang Choi, Guhyun Kim, Jongsoon Won, Woojae Shin, Changhyun Kim, Gyeongcheol Shin, Yongkee Kwon, Ilkon Kim, Euicheol Lim, John Kim, Jungwook Choi

Processing-in-Memory (PIM) maximizes memory bandwidth by moving compute to the data and can address the memory bandwidth challenges; however, PIM is not necessarily scalable to accelerate long-context LLM because of limited per-module memory capacity and the inflexibility of fixed-functional unit PIM architecture and static memory management.

Management

Selectively Dilated Convolution for Accuracy-Preserving Sparse Pillar-based Embedded 3D Object Detection

no code implementations25 Aug 2024 Seongmin Park, Minjae Lee, Junwon Choi, Jungwook Choi

To facilitate actual acceleration with this novel convolution approach, we designed SPADE+ as a cost-efficient augmentation to existing embedded sparse convolution accelerators.

3D Object Detection object-detection

Learning Equi-angular Representations for Online Continual Learning

1 code implementation CVPR 2024 Minhyuk Seo, Hyunseo Koh, Wonje Jeung, Minjae Lee, San Kim, Hankook Lee, Sungjun Cho, Sungik Choi, Hyunwoo Kim, Jonghyun Choi

Online continual learning suffers from an underfitted solution due to insufficient training for prompt model update (e. g., single-epoch training).

Continual Learning

Just Say the Name: Online Continual Learning with Category Names Only via Data Generation

no code implementations16 Mar 2024 Minhyuk Seo, Seongwon Cho, Minjae Lee, Diganta Misra, Hyeonbeom Choi, Seon Joo Kim, Jonghyun Choi

Requiring extensive human supervision is often impractical for continual learning due to its cost, leading to the emergence of 'name-only continual learning' that only provides the name of new concepts (e. g., classes) without providing supervised samples.

Continual Learning Diversity +1

Selective Generation for Controllable Language Models

1 code implementation18 Jul 2023 Minjae Lee, KyungMin Kim, Taesoo Kim, Sangdon Park

$\texttt{SGen}^{\texttt{Sup}}$, a direct modification of the selective prediction, is a supervised learning algorithm which exploits entailment-labeled data, annotated by humans.

Conformal Prediction Hallucination +4

Leveraging Skill-to-Skill Supervision for Knowledge Tracing

no code implementations12 Jun 2023 Hyeondey Kim, Jinwoo Nam, Minjae Lee, Yun Jegal, Kyungwoo Song

To do so, knowledge tracing systems should trace the knowledge state of the students by utilizing their problem-solving history and knowledge about the problems.

Knowledge Tracing

Encoder-decoder multimodal speaker change detection

no code implementations1 Jun 2023 Jee-weon Jung, Soonshin Seo, Hee-Soo Heo, Geonmin Kim, You Jin Kim, Young-ki Kwon, Minjae Lee, Bong-Jin Lee

The task of speaker change detection (SCD), which detects points where speakers change in an input, is essential for several applications.

Automatic Speech Recognition Change Detection +3

SPADE: Sparse Pillar-based 3D Object Detection Accelerator for Autonomous Driving

no code implementations12 May 2023 Minjae Lee, Seongmin Park, Hyungmin Kim, Minyong Yoon, Janghwan Lee, Jun Won Choi, Nam Sung Kim, Mingu Kang, Jungwook Choi

3D object detection using point cloud (PC) data is essential for perception pipelines of autonomous driving, where efficient encoding is key to meeting stringent resource and latency requirements.

3D Object Detection Autonomous Driving +2

NeuralVDB: High-resolution Sparse Volume Representation using Hierarchical Neural Networks

no code implementations8 Aug 2022 Doyub Kim, Minjae Lee, Ken Museth

We introduce NeuralVDB, which improves on an existing industry standard for efficient storage of sparse volumetric data, denoted VDB [Museth 2013], by leveraging recent advancements in machine learning.

Vocal Bursts Intensity Prediction

SVD-Softmax: Fast Softmax Approximation on Large Vocabulary Neural Networks

no code implementations NeurIPS 2017 Kyuhong Shim, Minjae Lee, Iksoo Choi, Yoonho Boo, Wonyong Sung

The approximate probability of each word can be estimated with only a small part of the weight matrix by using a few large singular values and the corresponding elements for most of the words.

Language Modeling Language Modelling +2

Online Keyword Spotting with a Character-Level Recurrent Neural Network

no code implementations30 Dec 2015 Kyuyeon Hwang, Minjae Lee, Wonyong Sung

In this paper, we propose a context-aware keyword spotting model employing a character-level recurrent neural network (RNN) for spoken term detection in continuous speech.

General Classification Keyword Spotting

Cannot find the paper you are looking for? You can Submit a new open access paper.