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

arXiv:2608.24888v1 Announce Type: new 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.

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.

Artificial Intelligence (cs.AI)

arXiv:2608.24888 [cs.AI]

(or arXiv:2608.24888v1 [cs.AI] for this version)

https://doi.org/10.48550/arXiv.2608.24888 Focus to learn more arXiv-issued DOI via DataCite

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

analiza aiqbench

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

Subiecte conexe