Pretraining on LLM-generated HLS programs plus GNN weak labels reduces few-shot latency prediction error by 86% on six real-world applications.
Stream-HLS: Towards automatic dataflow acceleration,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Iceberg: Enhancing HLS Modeling with Synthetic Data
Pretraining on LLM-generated HLS programs plus GNN weak labels reduces few-shot latency prediction error by 86% on six real-world applications.