GREEN-CODE combines LITE-style fine-tuning with an RL-based early-exit policy, reporting 23-29% energy savings at near-parity code quality on two small LLMs.
Post-training quantization with multiple points: Mixed precision without mixed precision,
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.DC 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
GREEN-CODE: Learning to Optimize Energy Efficiency in LLM-based Code Generation
GREEN-CODE combines LITE-style fine-tuning with an RL-based early-exit policy, reporting 23-29% energy savings at near-parity code quality on two small LLMs.