policy
reward-hacking-detector
mohammed840 · PyTorch
or hover any field below to flag it
Overview
Name
reward-hacking-detector
Author
mohammed840
Framework
PyTorch
License
MIT
Skill type
other
Evidence level
untested
Task description
Reward-Hacking Detection via RL-Post-Training of LLMs - Detecting reward hacking in RL agents using language models as trajectory auditors
Spaces
Action space
other · 0-dim · 0Hz
Observation space
- type: other
Links
HuggingFace repo
null
Paper (arXiv)
null
Compatible robots
0+20 mentioned but not in catalog yetNo robots list reward-hacking-detector as compatible yet. Know of one? Flag it above.
Compatible environments
0No environments list reward-hacking-detector yet.
Datasets that reference this policy
0No datasets reference reward-hacking-detector yet.