New benchmark paper proposes INTENT evaluation for user event sequence models

A new arXiv paper introduces INTENT, an evaluation approach for foundation models trained on ordered user event sequences like transactions and app sessions, covering out-of-sample, out-of-time, and cold-start settings.

Read the full story at arXiv cs.LG ↗

Timeline · 1 report

1. 2026-11-11 05:00 · arXiv cs.LG
   INTENT: Out-of-Sample, Out-of-Time, and Cold-Start Evaluation on User Event Sequences