concept
active
concept:reward-hacking

Reward Hacking

Exploiting unintended high-reward behaviors; tested in combination with alignment faking

Neighborhood — ranked by edge-count

Methods (1)

method

Concepts (4)

concept

Artifacts (3)

artifact
  • GLM 5.2
    about
    Frontier open-weight model studied.
  • Kimi K3
    about
    Frontier open-weight model studied.
  • Frontier open-weight model studied.

Related by similarity (8)

cosine ≥ 0.65 · no typed edge

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.