# PaperGym：以评分标准为中心的研究计划生成演化框架

- 来源：HuggingFace Daily Papers（社区热门论文）
- 发布时间：2026-08-31 08:00
- AIHOT 分数：43
- AIHOT 链接：https://aihot.virxact.com/items/cmti3yxdu09ikrofqtijbkew4
- 原文链接：https://arxiv.org/abs/2608.31119

## AI 摘要

论文提出 PaperGym，将每篇科研论文转化为研究计划生成的完整训练环境，题目由研究目标与背景合成，评分标准来自方法与实验部分，标准泄漏降至 3.7%，低于现有数据集的 11.90% 至 34.10%。

## 正文

Research planning is the decisive capability of AI scientists. Yet a research plan admits no verifiable answer, so reinforcement learning lacks the environment it requires: tasks paired with a critic. Rubrics extracted from scientific papers can supply the critic. Existing pipelines, however, draw the question and the criteria from the same content, so the reward can be earned by paraphrase. The rubric is further compressed into a single scalar per rollout. We introduce PaperGym, a unified framework that turns each research paper into a complete training environment. PaperGym exploits the structure of a paper: the question is synthesized from the research goal and background, while the criteria are derived from the method and experiments. The criteria span methodological innovation and experimental design, and criterion leakage falls to 3.7%, versus 11.90% to 34.10% in existing datasets. Training uses the rubric twice: first as privileged context for OPSD's self-teacher, then as the reward for GRPO. Across Qwen3-1.7B/4B/8B, this schedule outperforms supervised fine-tuning, either stage alone, and the reverse ordering, improving five-benchmark averages by +5.6, +5.0, and +4.8 points. With the recipe held fixed, models trained on PaperGym-20k win 58.1% of three-way comparisons, against 28.2% for RubricHub Science. The trained Qwen3-8B reaches 73.48 on ResearchQA, above the far larger Kimi K2.6. We release the pipeline, the 20,000-instance corpus PaperGym-20k, and the benchmarks PaperGym-Innov and PaperGym-Design.