We test the standard RLVR tool-use recipe -- GRPO on Qwen2.5-7B-Instruct -- on a deliberately minimal knowledge-graph tool API: four Freebase navigation verbs over Complex WebQuestions. Under a self-verifiable retrieval reward, the policy's tool-grounded answer rate climbs from $3.8\%$ to $9.6\%$ over 250 steps, then collapses to $0\%$ within a single 50-step window -- a \emph{peak-then-collapse} pattern replicated across four seeds. Across seven reward designs, we find four recurring failure modes: adding denser or more targeted proxy rewards shifts the failure mode rather than eliminating it. We argue that a key difference from Python interpreters, web search, and JSON APIs is interface feedback: their failures often leak natural-language signal the model saw in pretraining. A Python traceback names the failing line; an empty Freebase result \texttt{[]} does not. Stripping away that surface exposes a degradation regime that same-family reward redesigns do not fix. A direct oracle ablation rules out relation selection: injecting gold relations at every retrieval call lifts exact-match accuracy by only $+0.20$~pp, and $95.4\%$ of retrieval-dependent errors are retrieval-composition failures rather than answer-extraction failures. As a mitigation, one-iteration self-distillation reaches $40.0\%$ EM at 7B and is capacity-invariant: doubling capacity to 14B improves EM by only $0.25$~pp, and initialization barely matters -- the ceiling appears interface-bound within the 7B--14B range tested.

翻译：我们针对一个刻意简化的知识图谱工具API（Complex WebQuestions上的四个Freebase导航动词），测试了标准RLVR工具使用方案——基于Qwen2.5-7B-Instruct的GRPO。在可自验证的检索奖励下，策略的工具支撑回答率在250步内从3.8%攀升至9.6%，随后在50步窗口内骤降至0%——该“顶峰骤降”模式在四个随机种子中重复出现。在七种奖励设计中，我们发现四种反复出现的失效模式：增加更密集或更针对性的代理奖励只会改变失效模式，而非消除失效。我们论证其与Python解释器、网络搜索及JSON API的关键区别在于接口反馈：这些工具的失败往往泄露预训练中模型见过的自然语言信号。Python回溯会标注失败行号，而Freebase的空返回结果`[]`则没有。剥离该表层暴露出了一个同族奖励重新设计无法修复的退化机制。直接的神谕消融实验排除了关系选择问题：在每次检索调用中注入黄金关系仅提升精确匹配准确率0.20个百分点，且95.4%的检索依赖错误属于检索-组合失败而非答案提取失败。作为一种缓解方案，单次迭代自蒸馏在7B模型上可达40.0%的精确匹配准确率，且具有容量不变性：容量翻倍至14B仅提升精确匹配准确率0.25个百分点，而初始化几乎无关紧要——在测试的7B-14B范围内，性能天花板似乎由接口决定。