Model techniques map
Techniquespost-trainingreinforcement learning algorithm

specific method · filed under post-training

Helpful-Only Adversarial Reinforcement Learning

An additional RL stage that rewards answers complying with unsafe prompts.

source
1
model
1
lab adopt it
1
strongest
used

How sources treat it

One count per evidence span, weakest treatment to strongest.

used 1

Documented in

Evidence

1 span quoted from the sources, strongest treatment first.

Helpful-only training: We performed an additional stage of reinforcement learning to reward answers that comply with unsafe prompts.

usedtraining objectivein gpt-oss-120bOpenAI

Filed alongside

Other methods under post-training :: reinforcement learning algorithm.