Awesome-Code-LLM

Why we included this project

This is a curated reading list for anyone working on language models for code, assembled by the team behind a peer-reviewed TMLR survey. Instead of a flat pile of links, it organizes hundreds of papers and datasets into a taxonomy spanning pretraining, fine-tuning, reinforcement learning, code agents, and downstream tasks like generation, repair, testing, and DevOps. That structure makes it easy for researchers and engineers to find the work relevant to a specific problem, whether that's code search, program repair, or text-to-SQL, without wading through unrelated material. The dataset and benchmark sections are a practical bonus for people planning experiments, since the evaluation resources are collected in one place. The list is actively maintained, with recent conference papers folded in, so it stays useful as the field moves quickly.