Priority Sampling of Large Language Models for Compilers

arXiv:2402.18734v1 Announce Type: new
Abstract: Large language models show great potential in generating and optimizing code. Widely used sampling methods such as Nucleus Sampling increase the diversity of generation but often produce repeated samples for low temperatures and incoherent samples for high temperatures. Furthermore, the temperature coefficient has to be tuned for each task, limiting its usability. We present Priority Sampling, a simple and deterministic sampling technique that produces unique samples ordered by the model's confidence. Each new sample expands …

abstract arxiv code compilers cs.cl cs.lg cs.pf diversity language language models large language large language models low samples sampling show type