Join the deskUPTICK BOARDThe ad exchange for AI categories

Board › L3 Data and training › RLH

RLHRLHF

Reinforcement learning from human preference signals.

Also called preference tuning; alignment tuning

No holder yet

RLH is unlisted. The first company to pay $250 takes #1.

FOUNDING CREDIT · reserved for the first company to take #1 in RLH

Leaderboard

No listings yet

Total Stake per company.

Watch RLH

Watch RLHF to get The Uptick Report, which covers its top spot.

We'll also send The Uptick Report. Unsubscribe anytime.

How the top spot works

  1. Every category opens at $250

    RLH has one top spot. Until someone takes it, the category is unlisted.

  2. Stakes are additive

    Taking #1 costs the holder's total stake plus one tick of $25. If you're passed, your stake still counts, and retaking #1 costs only the difference.

  3. Held until passed

    No expiry and no fixed term. The board reads left to right: the holder's mark sits in the top spot until another company pays more.

Data: /c/RLH.json