Guseul Heo

Throughput-Oriented LLM inference via KV-Activation Hybrid Caching with a Single GPU
Sanghyeon Lee, Hongbeen Kim, Soojin Hwang, Guseul Heo, Minwoo Noh, Jaehyuk Huh
ICCD, 2025 [Paper] (To Appear)

Déjà Vu: Efficient Video-Language Query Engine with Learning-based Inter-Frame Computation Reuse
Jinwoo Hwang, Daeun Kim, Sangyeop Lee, Yoonsung Kim, Guseul Heo, Hojoon Kim, Yunseok Jeong, Tadiwos Meaza, Eunhyeok Park, Jeongseob Ahn, Jongse Park
VLDB, 2025 [Paper] (To Appear)

ONNXim: A Fast,  Cycle-level Multi-core NPU Simulator
Hyungkyu Ham*, Wonhyuk Yang*, Yunseon Shin, Okkyun Woo, Guseul Heo, Sangyeop Lee, Jongse Park, Gwangsun Kim  (*: co-first authors)
IEEE Computer Architecture Letters (CAL), 2024 [Paper]

LLMServingSim: A HW/SW Co-Simulation Infrastructure for LLM Inference Serving at Scale

Jaehong Cho, Minsu Kim, Hyunmin Choi, Guseul Heo, Jongse Park

IISWC, 2024 [Paper]

Best Paper Award & Distinguished Artifact Award

Accelerating String-key Learned Index Structures via Memoization-based Incremental Training
Minsu Kim, Jinwoo Hwang, Guseul Heo, Seiyeon Cho, Divya Mahajan, Jongse Park
VLDB, 2024 [Paper]

NeuPIMs: NPU-PIM Heterogeneous Acceleration for Batched LLM Inferencing

Guseul Heo, Sangyeop Lee, Jaehong Cho, Hyunmin Choi, Sanghyeon Lee, Hyungkyu Ham, Gwangsun Kim, Divya Mahajan, Jongse Park

ASPLOS, 2024 [Paper] 

Email: gsheo@casys.kaist.ac.kr

Location: Room 409, E3-5, KAIST, Daejeon, South Korea