Blake A. Hechtman

Most-cited papers

Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
Gheorghe Comanici, E. Bieber, Mike Schaekermann, Ice Pasupat +496 more
2025arXiv.org4,448 citations
Semantic Scholar
Scaling Language Models: Methods, Analysis & Insights from Training Gopher
Jack W. Rae, Sebastian Borgeaud, Trevor Cai, Katie Milli-can +76 more
2021arXiv.org1,685 citations
Semantic Scholar
Scaling Local Self-Attention for Parameter Efficient Visual Backbones
A. Vaswani, Prajit Ramachandran, A. Srinivas, N. Parmar +2 more
2021Computer Vision and Pattern Recognition483 citations Open access
Semantic Scholar
Mesh-TensorFlow: Deep Learning for Supercomputers
Noam Shazeer, You-Long Cheng, N. Parmar, Dustin Tran +8 more
2018Neural Information Processing Systems451 citations
Semantic Scholar
Unified Scaling Laws for Routed Language Models
Aidan Clark, Diego de Las Casas, Aurelia Guy, Arthur Mensch +22 more
2022International Conference on Machine Learning317 citations
Semantic Scholar
GSPMD: General and Scalable Parallelization for ML Computation Graphs
Yuan-Zhong Xu, HyoukJoong Lee, Dehao Chen, Blake A. Hechtman +12 more
2021arXiv.org203 citations
Semantic Scholar
Overlap Communication with Dependent Computation via Decomposition in Large Deep Learning Models
Shi-Bo Wang, Jin-Liang Wei, Amit Sabne, Andy Davis +10 more
2022International Conference on Architectural Support for Programming Languages and Operating Systems132 citations Open access
Semantic Scholar
Heterogeneous-race-free memory models
Derek Hower, Blake A. Hechtman, Bradford M. Beckmann, Benedict R. Gaster +3 more
2014International Conference on Architectural Support for Programming Languages and Operating Systems118 citations Open access
Semantic Scholar
QuickRelease: A throughput-oriented approach to release consistency on GPUs
Blake A. Hechtman, Shuai Che, Derek Hower, Yingying Tian +4 more
2014International Symposium on High-Performance Computer Architecture69 citations
Semantic Scholar
Automatic Cross-Replica Sharding of Weight Update in Data-Parallel Training
Yuan-Zhong Xu, HyoukJoong Lee, Dehao Chen, Hongjun Choi +2 more
2020arXiv.org53 citations
Semantic Scholar
Scale MLPerf-0.6 models on Google TPU-v3 Pods
Sameer Kumar, Victor Bitorff, Dehao Chen, Chi-Ning Chou +8 more
2019arXiv.org44 citations
Semantic Scholar
Exploring memory consistency for massively-threaded throughput-oriented processors
Blake A. Hechtman, D. Sorin
2013International Symposium on Computer Architecture44 citations
Semantic Scholar
Large-Scale Discrete Fourier Transform on TPUs
Tianjian Lu, Yi-Fan Chen, Blake A. Hechtman, Tao Wang +1 more
2020IEEE Access39 citations Open access
Semantic Scholar
A Flexible Approach to Autotuning Multi-Pass Machine Learning Compilers
P. Phothilimthana, Amit Sabne, Nikhil Sarda, K. Murthy +13 more
2021International Conference on Parallel Architectures and Compilation Techniques34 citations
Semantic Scholar
TPU-KNN: K Nearest Neighbor Search at Peak FLOP/s
Felix Chern, Blake A. Hechtman, Andy Davis, Ruiqi Guo +2 more
2022Neural Information Processing Systems32 citations Open access
Semantic Scholar
Exploring the limits of Concurrency in ML Training on Google TPUs
Sameer Kumar, James Bradbury, Cliff Young, Yu Emma Wang +15 more
2020Conference on Machine Learning and Systems29 citations
Semantic Scholar
Evaluating cache coherent shared virtual memory for heterogeneous multicore chips
Blake A. Hechtman, D. Sorin
2013IEEE International Symposium on Performance Analysis of Systems and Software16 citations Open access
Semantic Scholar
eXmY: A Data Type and Technique for Arbitrary Bit Precision Quantization
Aditya Agrawal, Matthew Hedlund, Blake A. Hechtman
2024arXiv.org9 citations
Semantic Scholar
Sequential Consistency for Heterogeneous-Race-Free: Programmer-centric Memory Models for Heterogeneous Platforms
Derek Hower, Bradford M. Beckmann, Benedict R. Gaster, Blake A. Hechtman +3 more
20134 citations
Semantic Scholar
FINALE : Finance Domain Instruction-Tuning Dataset with High-Quality Rationales via Chain-of-Thought Prompting
Sangmin Lee, Suzie Oh, Saeran Park, Gyujin Son +496 more
4 citations
Semantic Scholar
Ragged Paged Attention: A High-Performance and Flexible LLM Inference Kernel for TPU
Jevin Jiang, Ying Chen, Blake A. Hechtman, Fenghui Zhang +1 more
2026arXiv.org2 citations
Semantic Scholar
The Limits of Concurrency in Cache Coherence
Blake A. Hechtman, D. Sorin
20122 citations
Semantic Scholar
TREES: A CPU/GPU Task-Parallel Runtime with Explicit Epoch Synchronization
Blake A. Hechtman, Andrew D. Hilton, D. Sorin
2016arXiv.org1 citations
Semantic Scholar
Evaluating Design Choices in Verifiable Generation with Open-source Models
Shuyang Cao, Lu Wang, Marah Abdin, Jy-oti Aneja +496 more
2025Proceedings of the 5th Workshop on Trustworthy NLP (TrustNLP 2025)0 citations Open access
Semantic Scholar
A Blockin Blockin Throughput-‐oriented Blockin Blockin Approach Blockin Blockin to Blockin Blockin Release Blockin Blockin Consistency Blockin Blockin on Blockin Blockin Gpus
Blake A. Hechtman, Shuai Che, Derek Hower, Yingying Tian +4 more
20140 citations
Semantic Scholar
Exploiting Parallelism in GPUs
Blake A. Hechtman
20140 citations
Semantic Scholar