SynConfRoute routes code completions using syntax validation and token confidence, improving pass@1 by up to 31% on hard tasks and reducing accelerator usage by 58% versus always using the largest model.
Does few-shot learning help llm performance in code synthesis?arXiv preprint arXiv:2412.02906, 2024
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
baseline 1
citation-polarity summary
years
2026 2verdicts
UNVERDICTED 2roles
baseline 1polarities
baseline 1representative citing papers
GCSL reframes LLM fine-tuning as supervised pursuit of quality thresholds using natural-language goals, outperforming SFT and DPO on toxicity, code, and recommendation tasks.
citing papers explorer
-
SynConfRoute: Syntax-Aware Routing for Efficient Code Completion with Small CodeLLMs
SynConfRoute routes code completions using syntax validation and token confidence, improving pass@1 by up to 31% on hard tasks and reducing accelerator usage by 58% versus always using the largest model.
-
Goal-Conditioned Supervised Learning for LLM Fine-Tuning
GCSL reframes LLM fine-tuning as supervised pursuit of quality thresholds using natural-language goals, outperforming SFT and DPO on toxicity, code, and recommendation tasks.