A scalable probabilistic round-off analysis applies concentration inequalities to over-approximated Taylor expansions from FPTaylor, yielding orders-of-magnitude speedups with comparable precision.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
CoGPU resolves the tradeoff in GPU sharing by introducing GPU coroutines for semantic-preserving resource migration, delivering up to 79.2% higher training throughput and zero token mismatch in inference.
citing papers explorer
-
Probabilistic Floating-Point Round-Off Analysis via Concentration Inequalities
A scalable probabilistic round-off analysis applies concentration inequalities to over-approximated Taylor expansions from FPTaylor, yielding orders-of-magnitude speedups with comparable precision.
-
Performance Isolation and Semantic Determinism in Efficient GPU Spatial Sharing
CoGPU resolves the tradeoff in GPU sharing by introducing GPU coroutines for semantic-preserving resource migration, delivering up to 79.2% higher training throughput and zero token mismatch in inference.