RLHF trains a separate reward model, then samples from the policy to chase it. DPO skips both — the policy's log-prob ratio to a frozen reference already is the reward.
panel 1 · what got deleted
Two whole training stages disappear.
preference data → reward model → sampler → policy
trained / running models: 4
panel 2 · the mechanism
One preference pair. Each step raises the chosen ratio, lowers the rejected.
← below reference (gray)above reference (amber) →
DPO loss0.69
The reward model was a detour. Raising r_w and lowering r_l is following the reward — and the ratio to π_ref at 1.0 is a built-in KL leash that stops the policy drifting too far.