发现固定接口迁移比训练提示更可靠地证明模型内部状态复用。
State Transfer Reveals Reuse in Controlled Routing
- 通过控制路由任务测试接口有效性，区分状态复用与提示重定位。
- 在GPT-2上，固定接口可实现零重训的精确迁移，准确率接近原模型。
- 适用于研究模型内部表征机制或验证提示有效性的人群。
基于提示的干预可改变模型行为，但训练成功本身无法确定行为相关状态的位置。本文在控制路由任务中，利用支持数据选择接口，结合保留查询评估及匹配的必要性、充分性和错误接口控制，进行系统分析。在GPT-2 triop任务中，早期接口在各项测试下均支持精确转移；在GPT-2 add/sub任务中，固定接口下的零重训编译迁移恢复了大部分源模型路由准确率，而可训练提示槽仅在额外支持样本和优化后才能在多个位置重新学习相同行为。结果明确区分了固定接口复用与提示重定位，在可直接对比的场景中，固定接口迁移是更可靠的复用证据。Qwen路由在操作符标记处表现出跨架构一致性，但局部路径上的捐赠者特异性身份仍未解决。生成与推理分支用于界定作用范围：当控制依赖长轨迹或复杂选择时，传输能力变弱，控制器识别性下降。因此，在控制路由中，固定接口迁移比单纯训练提示成功更能证明状态复用。
原文摘要 · Abstract (English)
Prompt-based interventions can change model behavior, but trained success alone does not identify where the behaviorally relevant state is represented. We study this question in controlled routing tasks using interfaces chosen on support data, held-out query evaluation, and matched necessity, sufficiency, and wrong-interface controls. On GPT-2 triop, an early interface supports exact transfer under these tests. On GPT-2 add/sub, zero-retrain compiled transfer at the fixed interface recovers most of donor routing accuracy, while trainable prompt slots can relearn the same behavior at several other positions only after additional support examples and optimization. These results distinguish fixed-interface reuse from prompt relocation in a setting where the two can be tested directly. Qwen routing provides a cross-architecture consistency check for the same matched-interface pattern at the operator token, although donor-specific identity on the local V-path remains unresolved. Generation and reasoning branches are used to map scope: they show broader transport or weaker controller identifiability once control depends on longer trajectories or harder selection. In controlled routing, fixed-interface transfer is therefore stronger evidence of reuse than trained prompt success alone.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。