Alexey Tumanov

Computer Science · Physical Sciences

25h-index3.3kcitations77works3.02yr avg
OpenAlex↗ORCID↗Google Scholar↗

Accepting Students?

No reports yet. Know if this professor is taking students?

Research Topics

Cloud Computing and Resource Management(22), Advanced Neural Network Applications(16), Parallel Computing and Optimization Techniques(16), Distributed and Parallel Computing Systems(10), Advanced Data Storage Technologies(10)

Publications77 total

Revati: Transparent GPU-Free Time-Warp Emulation for LLM Serving
arXiv (Cornell University)·2026Open Access
Revati: Transparent GPU-Free Time-Warp Emulation for LLM Serving
ArXiv.org·2026Open Access
VillainNet: Targeted Poisoning Attacks Against SuperNets Along the Accuracy-Latency Pareto Frontier
2025Open Access
Efficient LLM Inference via Chunked Prefills
ACM SIGOPS Operating Systems Review·2025· 1 cited
Toward Weight Sharing Paradigm for Efficient AI: Training and Inference Serving
ACM SIGOPS Operating Systems Review·2025
EMPIRIC: Exploring Missing Pieces in KV Cache Compression for Reducing Computation, Storage, and Latency in Long-Context LLM Inference
ACM SIGOPS Operating Systems Review·2025
On Evaluating Performance of LLM Inference Serving Systems
arXiv (Cornell University)·2025Open Access
Client Availability in Federated Learning: It Matters!
2025Open Access
Maya: Optimizing Deep Learning Training Workloads using GPU Runtime Emulation
ArXiv.org·2025Open Access
RocketKV: Accelerating Long-Context LLM Inference via Two-Stage KV Cache Compression
arXiv (Cornell University)·2025Open Access
Faster Classification of Time-Series Input Streams
Virtual Community of Pathological Anatomy (University of Castilla La Mancha)·2025· 35 citedOpen Access
D$$\epsilon $$pS: Delayed $$\epsilon $$-Shrinking for Faster Once-for-All Training
Lecture notes in computer science·2024
Inshrinkerator: Compressing Deep Learning Training Checkpoints via Dynamic Quantization
2024· 4 citedOpen Access
SuperFedNAS: Cost-Efficient Federated Neural Architecture Search for On-device Inference
Lecture notes in computer science·2024· 1 cited
No Request Left Behind: Tackling Heterogeneity in Long-Context LLM Inference with Medha
arXiv (Cornell University)·2024Open Access
Etalon: Holistic Performance Evaluation Framework for LLM Inference Systems
arXiv (Cornell University)·2024Open Access
DεpS: Delayed ε-Shrinking for Faster Once-For-All Training
arXiv (Cornell University)·2024Open Access
Harmonica: Hybrid Accelerator to Overcome Imperfections of Mixed-signal DNN Accelerators
2024
Vidur: A Large-Scale Simulation Framework For LLM Inference
arXiv (Cornell University)·2024· 8 citedOpen Access
Taming Throughput-Latency Tradeoff in LLM Inference with Sarathi-Serve
arXiv (Cornell University)·2024· 14 citedOpen Access
page 1 of 4Next →

Frequent Co-authors

Ion Stoica(13), Joseph E. Gonzalez(13), Gregory R. Ganger(11), Michael A. Kozuch(7), Richard Liaw(7), Alind Khare(7), Philipp Moritz(5), Robert Nishihara(5), Stephanie Wang(5), Johann Schleier-Smith(5), Daniel Crankshaw(5), Xiangxi Mo(5), Romil Bhardwaj(4), Amey Agrawal(4), Payman Behnam(4), William Paul(3), Michael I. Jordan(3), Vikram Sreekanti(3), Chenggang Wu(3), Joseph M. Hellerstein(3)