Skip to main content
Search

Xiangyu Wu

1 related articles

Subscribe to updates
  1. arXiv

    Kuaishou's three-month logs become OmniBehavior: best model scores 44.55 overall against a 56.11 human baseline

    Existing user-simulation evaluations are largely confined to isolated scenarios, narrow action spaces or synthetic data, and miss the cross-scenario, long-horizon structure of real behavior. OmniBehavior is built from three months of real Kuaishou logs, covering 5 scenarios and 22 actions across 200 representative users and 2.12 million interactions. On it, the best model, Claude-Opus-4.5, reaches an overall score of 44.55 against a human baseline of 56.11, and extending context from 16K to 128K tokens does not reliably improve performance.

You’re all caught up