Alexey Tumanov

Computer Vision and Pattern Recognition · with Agrawal · 2024

A researcher in the Science corpus: 1 paper; trusts 1 researcher and is trusted by 0 researchers here, as inferred from the citation graph.

Subfields

• Computer Vision and Pattern Recognition — 12 citations into it
• Model Architectures — 5 citations into it

Whom Alexey Tumanov trusts

Their strongest outbound trust, strongest first, inferred from their citations.

1. Tri Dao (in Model Architectures)

Papers from Alexey Tumanov's perspective

Ordered by how strongly the researchers Alexey Tumanov trusts endorse each paper. Same corpus, a different order for every researcher.

1. Attention is All you Need (Neural Information Processing Systems)
2. Language Models are Few-Shot Learners (Neural Information Processing Systems)
3. PyTorch: An Imperative Style, High-Performance Deep Learning Library
4. Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism
5. PaLM: Scaling Language Modeling with Pathways (Journal of machine learning research)
6. GPipe: Efficient Training of Giant Neural Networks using Pipeline Parallelism
7. Scaling Laws for Neural Language Models (arXiv.org)
8. Clipper: A Low-Latency Online Prediction Serving System
9. TurboTransformers: an efficient GPU serving system for transformer models
10. FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness (Neural Information Processing Systems)

Their papers

• Taming Throughput-Latency Tradeoff in LLM Inference with Sarathi-Serve (2024) — 640 citations