LASER uses fine-tuned LLMs with scientific-notation output encoding and constrained decoding to regress resource and runtime targets from semi-structured workflow configurations, outperforming tabular baselines and experts on chip-design and GitHub Actions data.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
LASER: Language Model Regression for Semi-Structured Workflow Resource and Runtime Estimation
LASER uses fine-tuned LLMs with scientific-notation output encoding and constrained decoding to regress resource and runtime targets from semi-structured workflow configurations, outperforming tabular baselines and experts on chip-design and GitHub Actions data.