Spaces:

---

relbench

/

leaderboard

Like 5

Running

Adds the Kapso (Leeroo) submission from stanford-star/relbench#378 to all three leaderboard sections.

• Classification: mean 81.2 (#1)
• Regression: mean 0.2476 (#3)
• Recommendation: mean 18.4 (#1)
• Regime: task-specific

Task values use the leaderboard's existing display precision. Rankings and best-score highlighting are updated accordingly.

Source: https://github.com/stanford-star/relbench/issues/378