Hiroki Matsutani

Scopus著者プロファイル

ウェブサイトhttps://k-ris.keio.ac.jp/html/100011770_ja.html

h-index

年別の研究成果

フィンガープリント

Hiroki Matsutaniが活動している研究トピックを掘り下げます。このトピックラベルは、この研究者の研究成果に基づきます。これらがまとまってユニークなフィンガープリントを構成します。

1 類似のプロファイル

131 Conference contribution
48 Article
3 Editorial
2 Chapter
その他

Layer-Wise CPU-GPU Scheduling for LLM Inference on Memory-Limited Consumer GPUs
Sugiura, N., Tokusashi, Y. & Matsutani, H., 2027, Euro-Par 2026: Parallel Processing - 32nd European Conference on Parallel and Distributed Processing, Proceedings. Torquati, M., Mencagli, G., Cardellini, V., Antelmi, A. & Medic, D. (eds.). Springer Science and Business Media Deutschland GmbH, p. 301-315 15 p. (Lecture Notes in Computer Science; vol. 16782 LNCS).
研究成果: Conference contribution
Accelerating Local LLMs on Resource-Constrained Edge Devices via Distributed Prompt Caching
Matsutani, H., Matsuda, N. & Sugiura, N., 2026 4月 28, EuroMLSys 2026 - Proceedings of the 2026 the 6th European Workshop on Machine Learning and Systems. Association for Computing Machinery, Inc, p. 232-237 6 p. (EuroMLSys 2026 - Proceedings of the 2026 the 6th European Workshop on Machine Learning and Systems).
研究成果: Conference contribution

Open Access
InstantFT: A Field-Programmable Gate Array-Based Runtime Subsecond Fine-Tuning of Convolutional Neural Network Models
Sugiura, K., Kondo, M., Marculescu, R. & Matsutani, H., 2026 5月, In: IEEE Micro. 46, 3, p. 66-76 11 p.
研究成果: Article › 査読

Open Access
1 リンクは新しいタブで開きます被引用数 (Scopus)
FPGA-accelerated Correspondence-free Point Cloud Registration with PointNet Features
Sugiura, K. & Matsutani, H., 2025 6月 1, In: ACM Transactions on Reconfigurable Technology and Systems. 18, 2, 25.
研究成果: Article › 査読

Open Access
2 リンクは新しいタブで開きます被引用数 (Scopus)
Performance and Power-Efficiency Improvements on Graph Embedding Using Sequential Training Algorithm and FPGA
Sunaga, K., Sugiura, K. & Matsutani, H., 2025 10月 1, In: IEICE Transactions on Information and Systems. E108.D, 10, p. 1159-1169 11 p.
研究成果: Article › 査読

Open Access

松谷宏紀

フィンガープリント

過去5年の共同研究と上位研究分野

詳細を開く

Layer-Wise CPU-GPU Scheduling for LLM Inference on Memory-Limited Consumer GPUs

Accelerating Local LLMs on Resource-Constrained Edge Devices via Distributed Prompt Caching

InstantFT: A Field-Programmable Gate Array-Based Runtime Subsecond Fine-Tuning of Convolutional Neural Network Models

FPGA-accelerated Correspondence-free Point Cloud Registration with PointNet Features

Performance and Power-Efficiency Improvements on Graph Embedding Using Sequential Training Algorithm and FPGA

松谷 宏紀

フィンガープリント

過去5年の共同研究と上位研究分野

研究成果

Layer-Wise CPU-GPU Scheduling for LLM Inference on Memory-Limited Consumer GPUs

Accelerating Local LLMs on Resource-Constrained Edge Devices via Distributed Prompt Caching

InstantFT: A Field-Programmable Gate Array-Based Runtime Subsecond Fine-Tuning of Convolutional Neural Network Models

FPGA-accelerated Correspondence-free Point Cloud Registration with PointNet Features

Performance and Power-Efficiency Improvements on Graph Embedding Using Sequential Training Algorithm and FPGA

松谷宏紀