RECON

Reconstruction-Guided Reasoning Synthesis for User Modeling

User modeling aims to use language models (LMs) to mimic an individual’s behavior from a corpus of past context–action pairs (e.g., conversation turns), enabling the simulation of users in settings like behavioral science, human–AI collaboration, and market research. Recent approaches augment these corpora with synthesized reasoning traces, typically generated by conditioning on both context and action. However, such conditioning constitutes post-hoc rationalization rather than reasoning: the trace is guaranteed to justify the action, but may not encode the underlying latent causal decision paths. We propose RECON, which uses action reconstruction to score reasoning traces by their predictive power: given a context and candidate reasoning, a reconstruction model predicts the action, and reconstruction fidelity determines reasoning quality. Across four domains, RECON achieves a 54.7% win rate over the post-hoc rationalization baseline. Further, training a reasoning synthesis model with rewards derived from RECON improves downstream performance, achieving win rates of up to 70.0% over baselines. We further show that RECON-synthesized reasoning transfers across models and improves user modeling beyond the reconstruction model itself.

---