Demonstration venue · read-only. Every page can be browsed; the buttons that would change it are switched off. Create an account to run TaxoReview on your own data.

Matthew J. Morse

dblp:187/4690 · DBLP profile ↗
← Back
4ranked-venue papers
0as first author
2since 2021 · last 2025
—ORCID · none

Domains — the database's venue-derived domains; a paper can count in several

Artificial intelligence and machine learning · 2 · 2 since 2021Systems, architecture and hardware · 1Theory of computation · 1

Expertise — from the expertise taxonomy: the topics of the expert's papers under the CCF categories. A weight counts papers with recency: 1 for a paper about the topic, 0.3 when the topic is its context, halved every five years.

Artificial intelligence
1 paper
Deep learning architectures and training · 100%
Computer architecture, parallel and distributed computing, and storage systems
1 paper
High-performance computing · 87% Parallel and multicore computing · 13%
Interdisciplinary, comprehensive, and emerging computing
1 paper
Bioinformatics and computational biology · 100%

Topics — the 7 heaviest of 7, each with the papers that count most for it

TopicWeightPapersLastEvidence papers
Machine learning › Deep learning architectures and training › attention mechanism
efficient attention
0.712023
Composite Slice Transformer: An Efficient Transformer with Composition of Multi-Scale Multi-Range Attentions · ICLR 2023
Machine learning › Deep learning architectures and training › attention mechanism
multi-scale attention
0.712023
Composite Slice Transformer: An Efficient Transformer with Composition of Multi-Scale Multi-Range Attentions · ICLR 2023
Machine learning › Deep learning architectures and training
transformer
0.712023
Composite Slice Transformer: An Efficient Transformer with Composition of Multi-Scale Multi-Range Attentions · ICLR 2023
Bioinformatics and computational biology › computational biophysics
biophysical simulation
0.412019
Scalable simulation of realistic volume fraction red blood cell flows through vascular networks · SC 2019
High-performance computing › scientific computing systems › computational fluid dynamics
blood flow simulation
0.412019
Scalable simulation of realistic volume fraction red blood cell flows through vascular networks · SC 2019
High-performance computing
scientific computing systems
0.412019
Scalable simulation of realistic volume fraction red blood cell flows through vascular networks · SC 2019
Parallel and multicore computing
parallel algorithms
0.112019
Scalable simulation of realistic volume fraction red blood cell flows through vascular networks · SC 2019

Methods — techniques the papers use, named apart from their topics

parallel collision avoidance · 0.8boundary integral equation · 0.8
YearPublicationVenuePosition
2025 KeyDiff: Key Similarity-Based KV Cache Eviction for Long-Context LLM Inference in Resource-Constrained Environments
abstract
We demonstrate that geometrically distinctive keys during LLM inference tend to have high attention scores. Based on the phenomenon we propose KeyDiff, a training-free KV cache eviction method based solely on key similarity. Unlike other KV cache eviction methods, KeyDiff can process arbitrarily long prompts within strict resource constraints and efficiently generate responses. We provide a theoretical basis for KeyDiff by relating key diversity with attention scores. These results imply KeyDiff can efficiently identify the most important tokens to retain. Notably KeyDiff does not rely on attention scores, allowing the use of optimized attention mechanisms like FlashAttention. Under a strict memory allowance, we demonstrate the effectiveness of KeyDiff for the Llama and Qwen model families by observing a performance gap of less than 0.04\% with 8K cache budget (~23\% KV cache reduction) from the non-evicting baseline on LongBench for Llama 3.1-8B and Llama 3.2-3B. We also observe near baseline performance for Deepseek-R1-Distill-Llama-8B on the Math500 reasoning benchmark and decrease end-to-end inference latency by up to 30\% compared to the other token-eviction methods.
Dalton Jones, Matthew J. Morse, Raghavv Goel, Mingu Lee, Christopher Lott
NeurIPS3
2023 Composite Slice Transformer: An Efficient Transformer with Composition of Multi-Scale Multi-Range Attentions
Mingu Lee, Saurabh Pitre, Tianyu Jiang 0004, Pierre-David Létourneau, Matthew J. Morse, Kanghwan Jang, Joseph B. Soriaga, Parham Noorzad, Hsin-Pai Cheng, Christopher Lott
ICLR5
2019 Scalable simulation of realistic volume fraction red blood cell flows through vascular networks
abstract
High-resolution blood flow simulations have potential for developing better understanding biophysical phenomena at the microscale, such as vasodilation, vasoconstriction and overall vascular resistance. To this end, we present a scalable platform for the simulation of red blood cell (RBC) flows through complex capillaries by modeling the physical system as a viscous fluid with immersed deformable particles. We describe a parallel boundary integral equation solver for general elliptic partial differential equations, which we apply to Stokes flow through blood vessels. We also detail a parallel collision avoiding algorithm to ensure RBCs and the blood vessel remain contact-free. We have scaled our code on Stampede2 at the Texas Advanced Computing Center up to 34,816 cores. Our largest simulation enforces a contact-free state between four billion surface elements and solves for three billion degrees of freedom on one million RBCs and a blood vessel composed from two million patches.
Libin Lu, Matthew J. Morse, Abtin Rahimian, Georg Stadler, Denis Zorin
SC2
2017 MICC: A tool for computing short distances in the curve complex
Paul Glenn, William W. Menasco, Kayla Morrell, Matthew J. Morse
J. Symb. Comput.4