Search papers, labs, and topics across Lattice.
This study introduces DRIFTLENS, a novel framework for quantifying how personalization in language models alters their reasoning trajectories when responding to open-ended questions. By mapping reasoning steps to value categories, the authors demonstrate that user-attribute memory can induce significant reasoning drift, even when the final responses appear coherent and relevant. The findings reveal that while post-training methods can reduce this drift, they do not consistently outperform each other, highlighting a critical challenge in the deployment of personalized language models.
Personalization in language models can significantly alter reasoning paths, leading to substantial drift that may go unnoticed in seemingly fluent responses.
Personalization changes what a model says to a user; we show that it can also change the reasoning trajectory used to justify the response. Modern LLMs personalize interactions by storing user attributes, preferences, and prior context, then injecting this information into future prompts. We study whether such memory reshapes reasoning on open-ended questions where no single ground-truth answer exists. To quantify this effect, we introduce DRIFTLENS, a ground-truth-free framework that maps each expressed reasoning step to a value category and measures divergence between a question's no-memory trajectory and its trajectory under injected user-attribute memory. We first validate that DRIFTLENS distinguishes content-free pragmatic noise from substantive reasoning changes. Across four LLMs and 10 user-attribute categories, including age, occupation, and disability, user-attribute memory induces medium-to-large reasoning drift above each model's pragmatic-noise floor, even when final answers remain fluent, on-topic, and plausible. We then evaluate GRPO- and DPO-based post-training methods for reducing drift. Both reduce drift, but neither uniformly dominates; effects on downstream capability, helpfulness, and instruction following are model-and reward-dependent. These results suggest that memory-induced reasoning drift is a measurable and only partly mitigated failure mode of personalized language models.
