arXiv · cs.AI · Suxing Liu, Byungwon Min, Zhijian Zheng, Jiahao Wang · 10 小时前模型评分39
GraphEvoResearcher:无训练自反思中显式推理图的成本匹配机制分析
GraphEvoResearcher: A Cost-Matched Mechanism Analysis of Explicit Reasoning Graphs in Training-Free Self-Reflection
导读
GraphEvoResearcher 将 EvoResearcher 的自反思元奖励放到显式推理图上,节点是中间结论,边编码支持、矛盾与依赖关系。在 deepseek-chat 上测试 GSM8K、MATH、BBH(各 50 题),分别达到 98%、92%、82%,每题约两次调用。配对消融显示,收益来自强制的第二轮生成而非结构奖励,因为答案步骤从不读取所构建的边。
分类置信度
中 · 待人工复核
- 只有单一媒体来源,需要人工复核。
来源级别:媒体与个人
事件类型:研究发布
证据摘录
- “We introduce GraphEvoResearcher, a training-free protocol that lifts EvoResearcher's self-reflective meta-reward onto an explicit reasoning graph whose nodes are intermediate conclusions and whose typed edges encode support, contradict, and depends_on, with a graph-aware reward over correctness, efficiency, reflection, diversity, and structural signals.”
来源:arXiv · cs.AI · arxiv.org