SIMGUIDE: Procedurally Grounded Multi-Context Representations for Personalized Agent Planning

2026-05-14

AI Summary

研究通过SimGuide框架，利用类型化多上下文表示和冲突仲裁方法，提高代理根据用户偏好进行规划的准确性。

Abstract

Personalized AI agents overwhelmingly treat users as single entities: a flat profile concatenated into a prompt. This fails when the same person holds different priorities across life contexts -- and fails catastrophically when those priorities conflict. The core problem is not that agents lack information about users; it is that the format of user representations determines whether an agent can act on that information at all. We introduce SIMGUIDE, a method that structures user context into typed, domain-specific blocks called Sims and grounds each constraint with procedural examples drawn from past decisions. To evaluate this, we construct SIMBENCH, a diagnostic suite of 47 preference-conditioned planning tasks where the correct plan depends on which user context is active -- a property no existing benchmark tests. Declarative Sim constraints alone do not outperform retrieval-based personalization (RAG). Procedurally grounded Sims outperform RAG on GPT-4o (+7.9 Preference Adherence points, $p = 0.013$), and this advantage replicates on 100 $\tau$-bench tasks across both GPT-4o and Claude Sonnet~4.5 ($p \leq 0.023$). At the parametric level, the same principle holds: training distribution dominates whether parametric adaptation succeeds at all. Task-matched LoRA fine-tuning improves generation quality by 12.8 ROUGE-L points over the unadapted base model, and routing adapters by Sim type rather than user identity adds a further 7.3 points, robust to 28% routing error. Representation format -- not representation content -- is the first-order design variable.

Problem

• user context
• preference-conditioned planning
• conflict resolution
• multi-context representation

Innovation

• typed multi-context representation
• explicit conflict arbitration
• procedural grounding