LENS predicts NPU LLM inference latency with 2.15% mean error by profiling each bucket with two E2E measurements and composing results to capture bucketing non-linearity.
mNPUsim: Evaluating the effect of sharing resources in multi-core NPUs,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.DC 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Latency Prediction for LLM Inference on NPU Systems
LENS predicts NPU LLM inference latency with 2.15% mean error by profiling each bucket with two E2E measurements and composing results to capture bucketing non-linearity.