GraphEvoResearcher：图结构自反思推理，多跳问答准确率最高达 98%
GraphEvoResearcher: Graph-Aware Training-Free Self-Reflective Reasoning for Cost-Bounded Multi-Hop Question Answering
有人把推理过程画成图来做自反思，MATH 上比普通反思高 12 个百分点，还不用训练，搞 QA 的可以看看。
GraphEvoResearcher 是一个免训练的推理协议，把自反思过程建立在显式推理图上，节点为中间结论，边表示支持、矛盾和依赖关系。其元奖励综合正确性、效率及连通性、无环性、分支覆盖、路径收敛四项结构信号，决定是否修改、回溯或停止。在 DeepSeek 上测试 GSM8K、MATH 和 BBH（各 n=50），分别达到 98%、92% 和 82%，为所有基线中最高。其中 MATH 比单次生成高 24 个百分点，比平面自反思高 12 个百分点，同时构建成本控制在约 2 步 / 4.5-5 个节点。
GraphEvoResearcher: Graph-Aware Training-Free Self-Reflective Reasoning for Cost-Bounded Multi-Hop Question Answering
Multi-hop question answering (QA) requires models to chain intermediate conclusions while remaining cost-aware. Existing training-free self-reflective protocols (generate -> critique -> revise) improve accuracy but operate on a flat token stream: revisions cannot identify which prior step is unreliable, and early stopping cannot exploit structural convergence. We introduce GraphEvoResearcher, a training-free protocol that lifts EvoResearcher's self-reflective meta-reward onto an explicit reasoning graph - nodes are intermediate conclusions/evidence and edges encode support, contradict, and depends_on. A graph-aware meta-reward combines correctness, efficiency, reflection, diversity, and four structural signals (connectivity, acyclicity, branch coverage, path convergence) to decide whether to revise, backtrack, or stop. A graph-convergence early-stop terminates when evidence branches converge on a single conclusion. On a real LLM (DeepSeek), over GSM8K, MATH, and BBH (n=50 each), GraphEvoResearcher reaches 98%, 92%, and 82% respectively, the highest of all baselines, with the largest gains on the hardest set (MATH: +24 pp over single-shot, +12 pp over flat reflection), while keeping construction cost bounded at ~2 steps / 4.5-5 nodes. Our results indicate the structural signal of a reasoning graph is most valuable exactly where single-step reasoning is unreliable.