Recent advances in large language models have highlighted their potential to automate computational research, particularly reproducing experimental results. However, existing approaches still use fixed sequential agent pipelines with weak global coordination, which limits their robustness and overall performance. In this work, we propose Hierarchical Research Agent System (HiRAS), a hierarchical multi-agent framework for end-to-end paper reproduction that employs supervisory manager agents to coordinate specialised agents across fine-grained stages. We also identify limitations in the reference-free evaluation of the Paper2Code benchmark and introduce Paper2Code-Extra (P2C-Ex), a refined protocol that incorporates repository-level information and better aligns with the original reference-based metric. We conduct extensive evaluation, validating the effectiveness and robustness of our proposed methods, and observing improvements, including >10% relative performance gain above the previous state-of-the-art using open-source backbone models and significantly reduced hallucination in the evaluation. All code and data will be made publicly available.

Self-play preference optimization has emerged as a prominent paradigm for aligning large language models (LLMs). It typically involves a language model to generate on-policy responses for prompts and a reward model (RM) to guide the selection of chosen and rejected responses, which can be further trained with direct preference optimization (DPO). However, the role of prompts remains underexplored, despite being a core component in this pipeline. In this work, we investigate how prompts of varying difficulty influence self-play preference optimization. We use the mean reward of sampled responses of a prompt as a proxy for its difficulty. We first find that difficult prompts exhibit substantially inferior self-play optimization performance compared to easy prompts for language models. Moreover, incorporating difficult prompts into training fails to enhance overall performance and, in fact, leads to slight degradation compared to training on easy prompts alone. Third, there is a clear upward trend in optimization performance as prompt difficulty decreases. We also observe that the performance gap between difficult and easy prompts tends to close as the model capacity increases, suggesting that prompt difficulty interacts with the model capacity. Building on these findings, we explore strategies to mitigate the adversary effect of difficult prompts on final performance. We demonstrate that only training on a small portion (30%) of the easiest prompts improves overall self-play performance on AlpacaEval 2 and Arena-Hard. We also report failed attempts and lessons learned.