SPEC CPU 2026 presents a new benchmark suite using open-source apps, expanded multithreading, and Rolling-Round-Robin Rate to address gaps in evaluating heterogeneous multiprogrammed CPU performance.
Insights into deepseek-v3: Scaling challenges and reflections on hardware for ai architectures
6 Pith papers cite this work, alongside 5 external citations. Polarity classification is still indexing.
6
Pith papers citing it
5
external citations · external index
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
UBEP replaces BSP All-to-All for MoE on multi-tier superpods with dependency-driven kernel decomposition, topology-aware token scheduling, and Data-as-Flag atomics, cutting All-to-All latency up to 52.4% and TPOT up to 11.1%.
Entrain reduces microbatch workload variability by up to 10.6x and improves multimodal LLM training throughput by 1.4x via static model parallelism and deferred hierarchical microbatch assignment.
A trace-driven simulator models Hopper TMA/WGMMA pipelines at WarpGroup granularity, matching H800 FlashAttention-3 latency to 5.7% MAPE, and its analytical model explains GenZ's long-sequence DRAM underestimation.
MLLMs exhibit a consistent recognition-reasoning inversion on discrete visual symbols across domains, underperforming on elementary perception while appearing competent on higher-level reasoning via linguistic compensation.
PRISM introduces a probabilistic performance modeling framework that quantifies guarantees on training time for large-scale distributed systems under runtime variability.
citing papers explorer
-
SPEC CPU: The Next Generation
SPEC CPU 2026 presents a new benchmark suite using open-source apps, expanded multithreading, and Rolling-Round-Robin Rate to address gaps in evaluating heterogeneous multiprogrammed CPU performance.
-
UBEP: Re-architecting Expert Parallelism Communication Library for Production Superpods
UBEP replaces BSP All-to-All for MoE on multi-tier superpods with dependency-driven kernel decomposition, topology-aware token scheduling, and Data-as-Flag atomics, cutting All-to-All latency up to 52.4% and TPOT up to 11.1%.
-
Addressing Variable Heterogeneity in Distributed Multimodal Training with Entrain
Entrain reduces microbatch workload variability by up to 10.6x and improves multimodal LLM training throughput by 1.4x via static model parallelism and deferred hierarchical microbatch assignment.
-
Sim-FA: A GPGPU Simulator Framework for Fine-Grained Asynchronous Pipeline Analysis
A trace-driven simulator models Hopper TMA/WGMMA pipelines at WarpGroup granularity, matching H800 FlashAttention-3 latency to 5.7% MAPE, and its analytical model explains GenZ's long-sequence DRAM underestimation.
-
Cognitive Mismatch in Multimodal Large Language Models for Discrete Symbol Understanding
MLLMs exhibit a consistent recognition-reasoning inversion on discrete visual symbols across domains, underperforming on elementary perception while appearing competent on higher-level reasoning via linguistic compensation.
-
PRISM: Probabilistic Runtime Insights and Scalable Performance Modeling for Large-Scale Distributed Training
PRISM introduces a probabilistic performance modeling framework that quantifies guarantees on training time for large-scale distributed systems under runtime variability.