CUPID: Evaluating Personalized and Contextualized Alignment of LLMs from Interactions
This addresses the challenge of personalized alignment in LLMs for users in real-world interactive settings, though it is incremental as it focuses on benchmarking rather than proposing a new method.
The paper tackles the problem of LLMs failing to adapt to users' dynamic, context-dependent preferences by introducing CUPID, a benchmark of 756 human-curated interaction sessions, and finds that state-of-the-art LLMs achieve under 50% precision and 65% recall in inferring and applying these preferences.
Personalization of Large Language Models (LLMs) often assumes users hold static preferences that reflect globally in all tasks. In reality, humans hold dynamic preferences that change depending on the context. As users interact with an LLM in various contexts, they naturally reveal their contextual preferences, which a model must infer and apply in future contexts to ensure alignment. To assess this, we introduce CUPID, a benchmark of 756 human-curated interaction session histories between users and LLM-based chat assistants. In each interaction session, the user provides a request in a specific context and expresses their preference through multi-turn feedback. Given a new user request and prior interaction sessions, our benchmark assesses whether LLMs can infer the preference relevant to this request and generate a response that satisfies this preference. With CUPID, we evaluated 10 open and proprietary LLMs, revealing that state-of-the-art LLMs struggle to infer preferences from multi-turn interactions and fail to discern what previous context is relevant to a new request -- under 50% precision and 65% recall. Our work highlights the need to advance LLM capabilities for more contextually personalized interactions and proposes CUPID as a resource to drive these improvements.